├── COPYRIGHT
├── README.md
├── TED_comment_annotations
├── 1
│ ├── .svn
│ │ ├── entries
│ │ └── text-base
│ │ │ ├── comment_test.txt.svn-base
│ │ │ ├── sentence_test.txt.svn-base
│ │ │ └── thomas.svn-base
│ ├── comment_test.txt
│ ├── complete
│ │ ├── .svn
│ │ │ ├── entries
│ │ │ └── text-base
│ │ │ │ ├── comment_test.txt.svn-base
│ │ │ │ └── sentence_test.txt.svn-base
│ │ ├── comment_test.txt
│ │ └── sentence_test.txt
│ └── sentence_test.txt
├── 2
│ ├── .svn
│ │ ├── entries
│ │ └── text-base
│ │ │ ├── comment_test.txt.svn-base
│ │ │ ├── ilja.svn-base
│ │ │ └── sentence_test.txt.svn-base
│ ├── comment_test.txt
│ ├── complete
│ │ ├── .svn
│ │ │ ├── entries
│ │ │ └── text-base
│ │ │ │ ├── comment_test.txt.svn-base
│ │ │ │ └── sentence_test.txt.svn-base
│ │ ├── comment_test.txt
│ │ └── sentence_test.txt
│ └── sentence_test.txt
├── 3
│ ├── .svn
│ │ ├── entries
│ │ └── text-base
│ │ │ ├── comment_test.txt.svn-base
│ │ │ ├── dimitri.svn-base
│ │ │ └── sentence_test.txt.svn-base
│ ├── comment_test.txt
│ ├── complete
│ │ ├── .svn
│ │ │ ├── entries
│ │ │ └── text-base
│ │ │ │ ├── comment_test.txt.svn-base
│ │ │ │ └── sentence_test.txt.svn-base
│ │ ├── comment_test.txt
│ │ └── sentence_test.txt
│ └── sentence_test.txt
├── 4
│ ├── .svn
│ │ ├── entries
│ │ └── text-base
│ │ │ ├── comment_test.txt.svn-base
│ │ │ ├── marc.svn-base
│ │ │ └── sentence_test.txt.svn-base
│ ├── comment_test.txt
│ ├── complete
│ │ ├── .svn
│ │ │ ├── entries
│ │ │ └── text-base
│ │ │ │ ├── comment_test.txt.svn-base
│ │ │ │ └── sentence_test.txt.svn-base
│ │ ├── comment_test.txt
│ │ └── sentence_test.txt
│ └── sentence_test.txt
├── 5
│ ├── .svn
│ │ ├── entries
│ │ └── text-base
│ │ │ ├── comment_test.txt.svn-base
│ │ │ ├── nesli.svn-base
│ │ │ └── sentence_test.txt.svn-base
│ ├── comment_test.txt
│ ├── complete
│ │ ├── .svn
│ │ │ ├── entries
│ │ │ └── text-base
│ │ │ │ ├── comment_test.txt.svn-base
│ │ │ │ └── sentence_test.txt.svn-base
│ │ ├── comment_test.txt
│ │ └── sentence_test.txt
│ └── sentence_test.txt
└── 6
│ ├── .svn
│ ├── entries
│ └── text-base
│ │ ├── comment_test.txt.svn-base
│ │ ├── sentence_test.txt.svn-base
│ │ └── subheim.svn-base
│ ├── comment_test.txt
│ ├── complete
│ ├── .svn
│ │ ├── entries
│ │ └── text-base
│ │ │ ├── comment_test.txt.svn-base
│ │ │ └── sentence_test.txt.svn-base
│ ├── comment_test.txt
│ └── sentence_test.txt
│ └── sentence_test.txt
├── bootstrapping.py
├── datasets
├── .svn
│ ├── entries
│ └── text-base
│ │ ├── emoticons.data.svn-base
│ │ ├── emoticons_patch.py.svn-base
│ │ ├── segmentation_corpus.data.svn-base
│ │ └── subjclueslen1-HLTEMNLP05.tff.svn-base
├── __init__.py
├── __init__.pyc
├── emoticons.data
├── emoticons_patch.py
├── emoticons_patch.pyc
├── segmentation_corpus.data
└── subjclueslen1-HLTEMNLP05.tff
├── hp_classifiers.py
├── images
└── bootstrap.png
├── lexicon.py
├── lexicon
├── negative-words.txt
└── positive-words.txt
├── pb_classifiers.py
├── polarity.py
├── pos.py
├── replacer.py
├── sentiment.py
├── stored
├── learned_patterns
├── lexicon
└── ubt_tagger.classifier
├── svm.py
└── terminal_colors.py
/README.md:
--------------------------------------------------------------------------------
1 | usent — The attached code is a Python implementation of a dictionary-based sentiment classification procedure which combines two different bootstrapping procedures, namely for subjectivity and polarity detection (as in [3, 4] respectively). The rule-based polarity classifier is an extension of the one that was presented in [5]. Moreover, /TED_comment_annotations folder contains the files of the human study we conducted on TED comment sentiment classification (with 6 human annotators).
2 |
3 | - E. Riloff and J. Wiebe. Learning extraction patterns for subjective expressions.
4 | In Proceedings of the 2003 conference on Empirical methods in natural language
5 | processing, 2003.
6 | - D. K. M Wiegand. Bootstrapping supervised machine-learning polarity classifiers with rule-based classification.
7 | In Proceedings of the ECAI-Workshop on Computational Approaches to Subjectivity and Sentiment Analysis, 2009.
8 | - T. Wilson, J. Wiebe, and P. Hoffmann. Recognizing contextual polarity in phrase-level sentiment
9 | analysis. In Proceedings of the conference on Human Language Technology and Empirical Methods in
10 | Natural Language Processing, 2005.
11 |
12 | The code was used for an opinion mining and retrieval system presented at CICLing 2013 [1], and for improving one-class collaborative filtering [2].
13 |
14 | ```
15 | @incollection{pappas13c,
16 | location = {Samos, Greece},
17 | year = {2013},
18 | booktitle = {Computational Linguistics and Intelligent Text Processing},
19 | volume = {7817},
20 | doi = {10.1007/978-3-642-37256-8_17},
21 | title = {Distinguishing the Popularity between Topics: A System for Up-to-Date Opinion Retrieval and Mining in the Web},
22 | author = {Pappas, Nikolaos and Katsimpras, Georgios and Stamatatos, Efstathios},
23 | pages = {197-209}
24 | }
25 |
26 | @inproceedings{pappas13a,
27 | author = {Pappas, Nikolaos and Popescu-Belis, Andrei},
28 | title = {Sentiment Analysis of User Comments for One-Class Collaborative Filtering Over {TED} Talks},
29 | booktitle = {Proceedings of the 36th international ACM SIGIR conference on Research and development in information retrieval},
30 | series = {SIGIR '13},
31 | year = {2013},
32 | isbn = {978-1-4503-2034-4},
33 | location = {Dublin, Ireland},
34 | pages = {773--776},
35 | numpages = {4},
36 | }
37 | ```
38 |
39 |
40 |
41 | Installing dependencies
42 | ------------
43 | The available code for unsupervised sentiment classification requires Python programming
44 | language and pip package manager to run. For detailed installing instructions please refer to
45 | the following links:
46 | http://www.python.org/getit/
47 | http://www.pip-installer.org/en/latest/
48 |
49 | After installing them, you should be able to install the following packages:
50 | ```bash
51 | $ pip install nltk
52 | $ pip install stemmer
53 | $ pip install numpy
54 | $ pip install pickle
55 | ```
56 |
57 | After you install nltk you will need some corpora to train the sequential POS tagger (pos.py) and the nltk tokenizer.
58 | ```bash
59 | $ python
60 | ```
61 | ```python
62 | import nltk
63 | nltk.download()
64 | ```
65 | The issue of the above command will load a graphical interface that lets you manage several corpora
66 | related to nltk library. From the list select and download the following corpora:
67 | *tokenizers/punkt/english*, *wordnet*, *brown*, *conll2000* and *treebank*.
68 |
69 | Lastly, pyml library is needed for the SVM classifier that is used currently in our code.
70 | Download http://pyml.sourceforge.net/ and then issue:
71 | ```bash
72 | $ tar zxvf PyML-0.7.11.tar.gz
73 | $ cd PyML-0.7.11
74 | $ python setup.py build
75 | $ python setup.py install
76 | ```
77 |
78 |
79 | Processing pipeline
80 | -------------------
81 | The current pipline that is implemented in sentiment.py is depicted in the following diagram. Initially,
82 | the input text is split into sentences and each sentence is fed to a high precision subjectivity classifier.
83 | If the sentence is classified as subjective then syntactic patterns are learned from this instance. In case
84 | that the sentence is not detected as such then it is fed to the pattern-based classifier. The pattern-based
85 | classifier outputs the class of the sentence based on the learned patterns so far. If the instance is subjective
86 | then again more patterns are learned from it, otherwise it is fed to a high precision objectivity classifier.
87 | If the sentence is classified as objective, then it is ignored, otherwise it is fed to the polarity classifier.
88 | Finally, the polarity classifier estimates the numerical sentiment and normalized sentiment values and outputs
89 | the result. The instances with high confidence from the polarity classifier can be further used to train an SVM
90 | classifier to improve further the classification performance (see paper for further details). At the current version
91 | this option is disabled, but you can easily enable it. Similarly, you can remove some of the components from the
92 | pipeline according to your needs (e.g. skip subjectivity classification).
93 |
94 |
95 |
96 |
97 |
98 | Examples
99 | --------
100 | To estimate the total sentiment and total normalized sentiment (as described in the papers),
101 | you can simply execute the sentiment.py file and give the desired block of text as an argument.
102 | Make sure that you escape symbols such as '"' and '!'. Apart from the command line execution you
103 | can integrate the library to your code and use directly the returned results. Below you can
104 | find two simple examples for demonstrating purposes:
105 |
106 | ```bash
107 | $ python sentiment.py "I have to give much love and respect to Rony. Your work is Amazing\\!"
108 | ```
109 | ```
110 | [+] Loaded existing UBT tagger!
111 | [+] Loaded existing pattern knowledge!
112 |
113 | [*] Checking block of text:
114 | [1] I have to give much love and respect to Rony.
115 | [2] Your work is Amazing!
116 |
117 | [*] Overall sentiment analysis:
118 |
119 | Parts: 2
120 | Sentiments: ['positive', 'positive']
121 | Scores: [4, 6.0]
122 | Results: {'positive': {'count': 2, 'score': 10.0, 'nscore': 1.9},
123 | 'neutral': {'count': 0, 'score': 0, 'nscore': 0},
124 | 'negative': {'count': 0, 'score': 0, 'nscore': 0}}
125 |
126 |
127 | subjective-----> 100.00%
128 | objective------> 0.00%
129 |
130 |
131 | positive-------> 100.00%
132 | neutral--------> 0.00%
133 | negative-------> 0.00%
134 |
135 | [x] positive (10.00, 1.90)
136 | ```
137 |
138 |
139 | ```bash
140 | $ python sentiment.py "I was blown away by some of the comments here posted by people who is either
141 | uneducated, ignorant, self-righteous or al-of-the-above. I'm irritated and saddened as I read these
142 | finger-pointing \"i'm right and you're wrong\" type of posts\!"
143 | ```
144 | ```
145 | [+] Loaded existing UBT tagger!
146 | [+] Loaded existing pattern knowledge!
147 |
148 | [*] Checking block of text:
149 | [1] I was blown away by some of the comments here posted by people who is either uneducated, ignorant, self-righteous or al-of-the-above.
150 | [2] I'm irritated and saddened as I read these finger-pointing "i'm right and you're wrong" type of posts!
151 |
152 | [*] Overall sentiment analysis:
153 |
154 | Parts: 2
155 | Sentiments: ['negative', 'negative']
156 | Scores: [-4, -4.0]
157 | Results: {'positive': {'count': 0, 'score': 0, 'nscore': 0},
158 | 'neutral': {'count': 0, 'score': 0, 'nscore': 0},
159 | 'negative': {'count': 2, 'score': -8.0, 'nscore': -0.3722943722943723}}
160 |
161 |
162 | subjective-----> 100.00%
163 | objective------> 0.00%
164 |
165 |
166 | positive-------> 0.00%
167 | neutral--------> 0.00%
168 | negative-------> 100.00%
169 |
170 | [x] negative (-8.00, -0.37)
171 | ```
172 |
--------------------------------------------------------------------------------
/TED_comment_annotations/1/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 2043
5 | svn://deepcore.gr/dc/ted/v1/hlt_project/utils/full_corpus/subjects/testing/1
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2013-01-18T18:08:40.681354Z
11 | 1662
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | comment_test.txt
30 | file
31 |
32 |
33 |
34 |
35 | 2013-01-18T19:54:08.000000Z
36 | 866d3f79f17831ae93d69b7a3e60fe63
37 | 2013-01-18T18:08:40.681354Z
38 | 1662
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 34783
62 |
63 | sentence_test.txt
64 | file
65 |
66 |
67 |
68 |
69 | 2013-01-18T19:54:08.000000Z
70 | 1f8e13fcdf3ae509424e4d068f5530cd
71 | 2013-01-18T18:08:40.681354Z
72 | 1662
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 14723
96 |
97 | complete
98 | dir
99 |
100 | thomas
101 | file
102 |
103 |
104 |
105 |
106 | 2013-01-18T19:54:08.000000Z
107 | d41d8cd98f00b204e9800998ecf8427e
108 | 2013-01-18T18:08:40.681354Z
109 | 1662
110 | snupdc
111 |
112 |
113 |
114 |
115 |
116 |
117 |
118 |
119 |
120 |
121 |
122 |
123 |
124 |
125 |
126 |
127 |
128 |
129 |
130 |
131 |
132 | 0
133 |
134 |
--------------------------------------------------------------------------------
/TED_comment_annotations/1/.svn/text-base/thomas.svn-base:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/TED_comment_annotations/1/.svn/text-base/thomas.svn-base
--------------------------------------------------------------------------------
/TED_comment_annotations/1/complete/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 2043
5 | svn://deepcore.gr/dc/ted/v1/hlt_project/utils/full_corpus/subjects/testing/1/complete
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2013-01-18T18:08:40.681354Z
11 | 1662
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | comment_test.txt
30 | file
31 |
32 |
33 |
34 |
35 | 2013-01-18T19:54:08.000000Z
36 | a427c44f4483e55f129daef2756b86fc
37 | 2013-01-18T18:08:40.681354Z
38 | 1662
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 34826
62 |
63 | sentence_test.txt
64 | file
65 |
66 |
67 |
68 |
69 | 2013-01-18T19:54:08.000000Z
70 | 3af629e72cc2189c0abe62e8e60e50dc
71 | 2013-01-18T18:08:40.681354Z
72 | 1662
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 14863
96 |
97 |
--------------------------------------------------------------------------------
/TED_comment_annotations/1/sentence_test.txt:
--------------------------------------------------------------------------------
1 | 1. No magic, alien assistance or lost mystical powers - just plain patience and determination ignoringg the attendant slavery for the moment).
2 | --->
3 |
4 | 2. Not only does he put focus on the dumbing down, dilution and perversion of scientific findings one sees presented in the media, but he explains the mechanisms by which the same findings become inaccessible or incomprehensible to the less probing masses.
5 | --->
6 |
7 | 3. This is science at it's best and the future is bright.
8 | --->
9 |
10 | 4. Is this really a good idea in the long term?
11 | --->
12 |
13 | 5. The inventor did not mean to produce such kind of glue.
14 | --->
15 |
16 | 6. People do not want to change now as long as oil is still affordable.
17 | --->
18 |
19 | 7. You fail to realize what povertyy" really means in regards to education, connection, status, class, family and social clasism.
20 | --->
21 |
22 | 8. clumsy to say the less) but I feel the full conclusion was not reached.
23 | --->
24 |
25 | 9. Agricultural techniques may increase that amount but doing so often comes with environmental damage and a loss of biodiversity.
26 | --->
27 |
28 | 10. Inspiring talk!
29 | --->
30 |
31 | 11. Think what you want, but evolution is true and there's no conspiracy.
32 | --->
33 |
34 | 12. The question of great power bringing great responsibility isn't new (spiderman, anyone?
35 | --->
36 |
37 | 13. With out ideologues, there is no hope whatsoever.
38 | --->
39 |
40 | 14. Beautiful, but for someone like me who's biggest problem is control"l over my experiences, it sounds like a sort of hell to not be able to see my boundaries!
41 | --->
42 |
43 | 15. Direction of Magnetic force, and 4.
44 | --->
45 |
46 | 16. i got a little bored listening.
47 | --->
48 |
49 | 17. Helping Handicapped and/or elderly people.
50 | --->
51 |
52 | 18. I do not mean to demean anyone.
53 | --->
54 |
55 | 19. I do love this concept of design for outcome, nice talk!
56 | --->
57 |
58 | 20. The most basic concept here is people thrive on autonomy.
59 | --->
60 |
61 | 21. The negative attitude is ingrained in us by evolution, no less.
62 | --->
63 |
64 | 22. Check out this newly created blog about bigg data", which in this wek's post discusses the ethical implications of this topic.
65 | --->
66 |
67 | 23. I love this!
68 | --->
69 |
70 | 24. I'm living in Japan, which consistently appears at the better end of the scale.
71 | --->
72 |
73 | 25. In his eloquent reminiscence of his visit to the Kaba- Islam's religious shrine, Mr. Mustafe Akyol uses the word, islamicc Law" several times.
74 | --->
75 |
76 | 26. Fantastic talk.
77 | --->
78 |
79 | 27. This sounds amazing.
80 | --->
81 |
82 | 28. More importantly, there is no government that would legitimize such businesses by issuing business permits or licenses, whose sole purpose is to police their role.
83 | --->
84 |
85 | 29. Aime is an extraordinary speaker.
86 | --->
87 |
88 | 30. I think it's pretty good to see someone stand out and endorse this kind of viewpoint providing a new way to see things.
89 | --->
90 |
91 | 31. I love her voice and the cello.
92 | --->
93 |
94 | 32. He doesn't have a clue about Functional Medicine and its uses in modern Europe.
95 | --->
96 |
97 | 33. Where's the innovation?
98 | --->
99 |
100 | 34. Looking at the data, I was struck not only by Thailand's rapid decrease in fertility rate but also by a similar decrease in fertility rate of its neighbors.
101 | --->
102 |
103 | 35. Many of my friends died in car accidents which could have been all avoided by this technology!
104 | --->
105 |
106 | 36. very nice work mr. klein.
107 | --->
108 |
109 | 37. Muchas veces al ver un problema pensamos que las soluciones siempre son complejas, lo cual implica de cierta forma m´s tiempo y m´s dinero; sin embargo existen soluciones simples y de bajo costo que permiten resultados similares, y eso es lo que vemos en este video que est´ orientado a las clases de escazos recursos.
110 | --->
111 |
112 | 38. the particles he mentions can be descriptions of different possible routs for vibrations of a very complex vortex of interacting stringss".
113 | --->
114 |
115 | 39. Not worthy of Ted.
116 | --->
117 |
118 | 40. not all of us may have the capital and resources that she had but i am pretty sure that most of us have the basic education needed to teach.
119 | --->
120 |
121 | 41. An amazing speaker with knowledge is enough to get me excited about any topic.
122 | --->
123 |
124 | 42. Most of all, it can inspire us to do things we never thought we could.
125 | --->
126 |
127 | 43. I believe that this debate has shown that wind for one can be a big part in solving our energy problems.
128 | --->
129 |
130 | 44. Bjorn reminds me of what I don't really like about organizations and events like Ted, Davos, the Cfr, etc.
131 | --->
132 |
133 | 45. I made the conscience decision to stop listening to music for a while which was devastating because music is my passion by far.
134 | --->
135 |
136 | 46. Many have denied in follow-up that they reject common descent (your 'macroevolution') and claim to have been misled or misrepresented.
137 | --->
138 |
139 | 47. The project requirements were to construct a minimal structure from unusual materials.
140 | --->
141 |
142 | 48. fascinating body of work and process of thinking
143 | --->
144 |
145 | 49. It is a kind of revolution in the area of technology.
146 | --->
147 |
148 | 50. @chris: Have you considered confirmation bias (chery-picking)?
149 | --->
150 |
151 | 51. Are you saying it's impossible for these apes to have a similar consciousness?
152 | --->
153 |
154 | 52. lastt phrase shamelessly copied from who knows who).
155 | --->
156 |
157 | 53. If anything, this lecture should have seen the different type.
158 | --->
159 |
160 | 54. But, its root cause, The Covert Enslavement of Humanity, is history's greatest injustice.
161 | --->
162 |
163 | 55. Dear T Walker sadlyy, you will not understand what that sentence really means".
164 | --->
165 |
166 | 56. Nick Bostrom; you can not prove a negative, self away beings will always search & need meaning, and from what we know from neuroscience, the brain seems to have evolved the ability to experience transcendence and belief.
167 | --->
168 |
169 | 57. This is a bad idea that leads to a generation or more that doesn't understand the fundamentals of calculation.
170 | --->
171 |
172 | 58. A brilliant professor !
173 | --->
174 |
175 | 59. Yet no one has been able to compare the effectiveness of solar orbital microwave power stations?
176 | --->
177 |
178 | 60. As far as his introduction, it was a hook and it was interesting and thought provoking, just like what followed.
179 | --->
180 |
181 | 61. Rahul Dewan: ii feel, the way to prevent terrorism is to 'raise the level of discourse' on God and Spirituality, and make it spread across all religious faiths".
182 | --->
183 |
184 | 62. And equally like the 19th century, our governing powers and ruling elites don't quite have any answer of how to undo what can't really be undone.
185 | --->
186 |
187 | 63. Commenting on Geof Mosher alongg with Tj Green) Iwould start b ystating the fact that evolutionn is war" is quite an understatment.
188 | --->
189 |
190 | 64. I will address them personally if you wish me to do so.
191 | --->
192 |
193 | 65. That's the wonderful right of freedom.
194 | --->
195 |
196 | 66. Very nice.
197 | --->
198 |
199 | 67. But could the opposite be true: that the gregariousness that is the driving force of human beings, is being virtualized by make-believe substitutes to actual social interaction?
200 | --->
201 |
202 | 68. Guys use Vlc player, the mp4 480p version works perfectly on it, awesome quality too.
203 | --->
204 |
205 | 69. Would it be impolite to ask where you stand in this argument?
206 | --->
207 |
208 | 70. Just wow.
209 | --->
210 |
211 | 71. However, had large numbers of Dutch with their hunting and military style rifles been able to start a sniping campaign against the Germans, Wi may have been shortened.
212 | --->
213 |
214 | 72. Amazing - Good Work - He is Worth to not only to India but also to the Whole world His promise for Open Source is also unbelivable a lots of Thanks to him and sponsers
215 | --->
216 |
217 | 73. It isn't good or bad; it's inevitable.
218 | --->
219 |
220 | 74. A couple of years ago the idea was about privacy that if you are connected to the Internet then your privacy is eliminated and you don't have any right to claim it back because you're a part of the network and you're obviously exchanging info that crosses a lot of paths to reach you, but recently, telecom providers are the ones dominating and watching, closely, what everyone is doing.
221 | --->
222 |
223 | 75. Wait for gasoline to hit $5 a gallon and it will.
224 | --->
225 |
226 | 76. Ib is a rigorous international education program for high school students.
227 | --->
228 |
229 | 77. As someone who has spent several years teaching, I can say from experience that children need discipline, a rigid structure and for the most part, they need to be told what to do.
230 | --->
231 |
232 | 78. Horrible architect, decent marketer perhaps?
233 | --->
234 |
235 | 79. They aren't reproducible on command, and they do not openly discussed the incidences because the subject is taboo.
236 | --->
237 |
238 | 80. Examples: To say that conservatives are much more into ideological purity is absurd.
239 | --->
240 |
241 | 81. David is talking about a job interview on steroids, and like all job interviews there is an understanding that things will be shown in the best light possible (to say the least).
242 | --->
243 |
244 | 82. In order to ensure we aren't fluctuations, we need to make an observation and analyze it.
245 | --->
246 |
247 | 83. They don’t have room or numbers and we are losing them.
248 | --->
249 |
250 | 84. The point I'm trying to make is this: a great craftsman is Not by default a great artist.
251 | --->
252 |
253 | 85. They are invited to raise their right hand and in unison say globall Truce Starts Now!
254 | --->
255 |
256 | 86. I would say all the fundamentalisms floating around in peoples minds making war on each other are one of the great problems of our time.
257 | --->
258 |
259 | 87. I will probably be one of the first to be swept away!
260 | --->
261 |
262 | 88. Please don't even Think of taking it down!
263 | --->
264 |
265 | 89. You see a lot of people, youngsters especially, who complain that they are depressed and alone.
266 | --->
267 |
268 | 90. I'm dubious about the payback periods you state.
269 | --->
270 |
271 | 91. Did his disgraceful comment on Israel fall on deaf ears?
272 | --->
273 |
274 | 92. To that end, how would empathy fix any of these problems?
275 | --->
276 |
277 | 93. follow his lead by taking a hard look at your own life and either get help or work on your own to become more 'gay'.
278 | --->
279 |
280 | 94. Sorry, he's a pet peeve.
281 | --->
282 |
283 | 95. human have no right and ability to make it!
284 | --->
285 |
286 | 96. Felix, why should an artist or a political activist not be commercially astute?
287 | --->
288 |
289 | 97. And now we have screwed up our heads were screwing about with monkeys heads as well):
290 | --->
291 |
292 | 98. It searches for and indexes common data in a vast resource and, not surprisingly, ends up finding lots of that data.
293 | --->
294 |
295 | 99. People need to try to be less offended by differing viewpoints.
296 | --->
297 |
298 | 100. their use of memetics suggests an intentionalityy" in memes.
299 | --->
300 |
301 | 101. A saying like childrenn are the pillars of Nation", at one time was just a saying without much ado.
302 | --->
303 |
304 | 102. Good stuff indeed, Dave.
305 | --->
306 |
307 | 103. Science can be used to gauge the value of some expressed morality, but&helip; only when one shares intention.
308 | --->
309 |
310 | 104. I am of the opinion that my life belongs to the community, and as long as I live, it is my privilege to do for it whatever I can.
311 | --->
312 |
313 | 105. Excellent speech.
314 | --->
315 |
316 | 106. ' 'how dare you say you know what we need, you don't know what it's like' 'emiliano's dad stole from me, so Emiliano should shut the hell up' ii am, of course, paraphrasing) what gives people the title of 'true mexican'?
317 | --->
318 |
319 | 107. For example - Lint Porter- " I can't confirm Steve Jones' numbers; but yes, it's safe to say that gay men are much more likely to be pedophiles.
320 | --->
321 |
322 | 108. As we continue to maintain the pace of technology development, it will be hard to take heed to what Dan Barber is preaching.
323 | --->
324 |
325 | 109. htp:/goglemapsapi.blogspot.com/208/01/put-your-map-on-map.html Now is that so hard Microsoft?
326 | --->
327 |
328 | 110. I found it interesting how Daniel used “think they’re hapy” to describe a person who moved from Ohio to California, but accepted the happiness as fact for those with incomes over $60,0.
329 | --->
330 |
331 | 111. I love the idea of technology that knows your emotional state and wants to help you, dims the lights for you when you're sleepy etc.
332 | --->
333 |
334 | 112. A hero can be someone who saves someone from making bad descisions or heading down the wrong path in their lives.
335 | --->
336 |
337 | 113. Really makes you think about genius and the creative process differently, at least for a moment.
338 | --->
339 |
340 | 114. Guess you're not ready for a public debate".
341 | --->
342 |
343 | 115. You're a real triumph who shows that things can be done.
344 | --->
345 |
346 | 116. Sending teachers and money to build school is a way more efficient long term way to fight terrorism then sending an army.
347 | --->
348 |
349 | 117. Pink is referring to were conducted, in part, by Dan Ariely, a Behavioral Economist from Mit, whose Ted talk addresses predictablyy irrational" human behavior.
350 | --->
351 |
352 | 118. Each year the garbage & recycling industry generates billions of dollars in revenue and recycling is the least of their priorities.
353 | --->
354 |
355 | 119. Thats completely honest.
356 | --->
357 |
358 | 120. Just before that, my 73 year old father had degenerative arthritis and needed a new hip to alleviate pain.
359 | --->
360 |
361 | 121. I suppose you consider them evil?
362 | --->
363 |
364 | 122. I find the fat-free, efficient part of the philosophy useful for handling and managing my work, schedule and projects.
365 | --->
366 |
367 | 123. And he who has deserved to drink from the ocean of life deserves to fill his cup from your little stream.
368 | --->
369 |
370 | 124. Cooked food, clothing, and the girl's room are part of modern comforts.
371 | --->
372 |
373 | 125. Facts about our happiness do not dictate moral conclusions, rather they must be interpreted with a number of moral principles in mind, such as e.g.
374 | --->
375 |
376 | 126. When this sensing device is attached to the sticky note, it senses whatever is written on this and transfer it to the Pc thereby making it a digital data.
377 | --->
378 |
379 | 127. I've been in a whel-chair for these 35 years and have never been sad or unhappy.
380 | --->
381 |
382 | 128. I just want a hall (150' long)?
383 | --->
384 |
385 | 129. Fascinating _
386 | --->
387 |
388 | 130. i am thinking now, thank you.
389 | --->
390 |
391 | 131. I really like the analogy of carbon atoms in two different objects totally having opposite properties and behaviours.
392 | --->
393 |
394 | 132. There is another option nobody seems to talk about because, what I think, a paradigm that is blocking vision: carr needs gas tank".
395 | --->
396 |
397 | 133. When we talk about costs, pricing, and profits, we need to remember who is the one paying for it.
398 | --->
399 |
400 | 134. Such a beautiful voice to deliver such a timely message as I get ready to top my last 'performance'.
401 | --->
402 |
403 | 135. I think you would too.
404 | --->
405 |
406 | 136. All the eccentricities aside, the main point about being passionate about learning came through.
407 | --->
408 |
409 | 137. There is so much real magic going on in the universe around us, I wish Ted would have focused on that instead of this.
410 | --->
411 |
412 | 138. But what a lot of you seem to not understand or simply ignore are that there are bad people out there that don't share your same values for life.
413 | --->
414 |
415 | 139. Despite the fact that a surprising proportion of people actually adhere to the current paradigm of what a healthy diet is, we are getting sicker and fatter despite mainstream dietary recommendations.
416 | --->
417 |
418 | 140. The only thing you'l find is a pacifier to suck or a security blanket to hold for those who want to believe there is life after death.
419 | --->
420 |
421 |
--------------------------------------------------------------------------------
/TED_comment_annotations/2/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 2043
5 | svn://deepcore.gr/dc/ted/v1/hlt_project/utils/full_corpus/subjects/testing/2
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2013-01-18T18:08:40.681354Z
11 | 1662
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | comment_test.txt
30 | file
31 |
32 |
33 |
34 |
35 | 2013-01-18T19:54:08.000000Z
36 | 6cf8c1dcb6f4c951027a6cadf32e786e
37 | 2013-01-18T18:08:40.681354Z
38 | 1662
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 31519
62 |
63 | ilja
64 | file
65 |
66 |
67 |
68 |
69 | 2013-01-18T19:54:08.000000Z
70 | d41d8cd98f00b204e9800998ecf8427e
71 | 2013-01-18T18:08:40.681354Z
72 | 1662
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 0
96 |
97 | sentence_test.txt
98 | file
99 |
100 |
101 |
102 |
103 | 2013-01-18T19:54:08.000000Z
104 | 011538bf103e354a0cc8d543bc0b4e71
105 | 2013-01-18T18:08:40.681354Z
106 | 1662
107 | snupdc
108 |
109 |
110 |
111 |
112 |
113 |
114 |
115 |
116 |
117 |
118 |
119 |
120 |
121 |
122 |
123 |
124 |
125 |
126 |
127 |
128 |
129 | 15078
130 |
131 | complete
132 | dir
133 |
134 |
--------------------------------------------------------------------------------
/TED_comment_annotations/2/.svn/text-base/ilja.svn-base:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/TED_comment_annotations/2/.svn/text-base/ilja.svn-base
--------------------------------------------------------------------------------
/TED_comment_annotations/2/complete/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 2043
5 | svn://deepcore.gr/dc/ted/v1/hlt_project/utils/full_corpus/subjects/testing/2/complete
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2013-01-18T18:08:40.681354Z
11 | 1662
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | comment_test.txt
30 | file
31 |
32 |
33 |
34 |
35 | 2013-01-18T19:54:08.000000Z
36 | df34d17361b67058e2323d38869318bf
37 | 2013-01-18T18:08:40.681354Z
38 | 1662
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 31561
62 |
63 | sentence_test.txt
64 | file
65 |
66 |
67 |
68 |
69 | 2013-01-18T19:54:08.000000Z
70 | 2c96232f4f3872f5dcba7a20f0ec02c8
71 | 2013-01-18T18:08:40.681354Z
72 | 1662
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 15218
96 |
97 |
--------------------------------------------------------------------------------
/TED_comment_annotations/3/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 2043
5 | svn://deepcore.gr/dc/ted/v1/hlt_project/utils/full_corpus/subjects/testing/3
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2013-01-18T18:08:40.681354Z
11 | 1662
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | comment_test.txt
30 | file
31 |
32 |
33 |
34 |
35 | 2013-01-18T19:54:08.000000Z
36 | 8a6af5675fa03e242fbd07128b65f48a
37 | 2013-01-18T18:08:40.681354Z
38 | 1662
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 30035
62 |
63 | dimitri
64 | file
65 |
66 |
67 |
68 |
69 | 2013-01-18T19:54:08.000000Z
70 | d41d8cd98f00b204e9800998ecf8427e
71 | 2013-01-18T18:08:40.681354Z
72 | 1662
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 0
96 |
97 | sentence_test.txt
98 | file
99 |
100 |
101 |
102 |
103 | 2013-01-18T19:54:08.000000Z
104 | 0d2e575a77990f2adda0594d795cb97b
105 | 2013-01-18T18:08:40.681354Z
106 | 1662
107 | snupdc
108 |
109 |
110 |
111 |
112 |
113 |
114 |
115 |
116 |
117 |
118 |
119 |
120 |
121 |
122 |
123 |
124 |
125 |
126 |
127 |
128 |
129 | 15069
130 |
131 | complete
132 | dir
133 |
134 |
--------------------------------------------------------------------------------
/TED_comment_annotations/3/.svn/text-base/dimitri.svn-base:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/TED_comment_annotations/3/.svn/text-base/dimitri.svn-base
--------------------------------------------------------------------------------
/TED_comment_annotations/3/complete/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 2043
5 | svn://deepcore.gr/dc/ted/v1/hlt_project/utils/full_corpus/subjects/testing/3/complete
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2013-01-18T18:08:40.681354Z
11 | 1662
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | comment_test.txt
30 | file
31 |
32 |
33 |
34 |
35 | 2013-01-18T19:54:08.000000Z
36 | b68a90f0fac3709583b5b4ba34c05289
37 | 2013-01-18T18:08:40.681354Z
38 | 1662
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 30095
62 |
63 | sentence_test.txt
64 | file
65 |
66 |
67 |
68 |
69 | 2013-01-18T19:54:08.000000Z
70 | 0c06884d6acff4d313b37a2625613440
71 | 2013-01-18T18:08:40.681354Z
72 | 1662
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 15209
96 |
97 |
--------------------------------------------------------------------------------
/TED_comment_annotations/4/.svn/entries:
--------------------------------------------------------------------------------
1 |
2 |
--------------------------------------------------------------------------------
/TED_comment_annotations/4/.svn/text-base/marc.svn-base:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/TED_comment_annotations/4/.svn/text-base/marc.svn-base
--------------------------------------------------------------------------------
/TED_comment_annotations/4/.svn/text-base/sentence_test.txt.svn-base:
--------------------------------------------------------------------------------
1 | 1. No magic, alien assistance or lost mystical powers - just plain patience and determination ignoringg the attendant slavery for the moment).
2 | --->
3 |
4 | 2. Not only does he put focus on the dumbing down, dilution and perversion of scientific findings one sees presented in the media, but he explains the mechanisms by which the same findings become inaccessible or incomprehensible to the less probing masses.
5 | --->
6 |
7 | 3. This is science at it's best and the future is bright.
8 | --->
9 |
10 | 4. Is this really a good idea in the long term?
11 | --->
12 |
13 | 5. The inventor did not mean to produce such kind of glue.
14 | --->
15 |
16 | 6. People do not want to change now as long as oil is still affordable.
17 | --->
18 |
19 | 7. You fail to realize what povertyy" really means in regards to education, connection, status, class, family and social clasism.
20 | --->
21 |
22 | 8. clumsy to say the less) but I feel the full conclusion was not reached.
23 | --->
24 |
25 | 9. Agricultural techniques may increase that amount but doing so often comes with environmental damage and a loss of biodiversity.
26 | --->
27 |
28 | 10. Inspiring talk!
29 | --->
30 |
31 | 11. Think what you want, but evolution is true and there's no conspiracy.
32 | --->
33 |
34 | 12. The question of great power bringing great responsibility isn't new (spiderman, anyone?
35 | --->
36 |
37 | 13. With out ideologues, there is no hope whatsoever.
38 | --->
39 |
40 | 14. Beautiful, but for someone like me who's biggest problem is control"l over my experiences, it sounds like a sort of hell to not be able to see my boundaries!
41 | --->
42 |
43 | 15. Direction of Magnetic force, and 4.
44 | --->
45 |
46 | 16. i got a little bored listening.
47 | --->
48 |
49 | 17. Helping Handicapped and/or elderly people.
50 | --->
51 |
52 | 18. I do not mean to demean anyone.
53 | --->
54 |
55 | 19. I do love this concept of design for outcome, nice talk!
56 | --->
57 |
58 | 20. The most basic concept here is people thrive on autonomy.
59 | --->
60 |
61 | 21. The negative attitude is ingrained in us by evolution, no less.
62 | --->
63 |
64 | 22. Check out this newly created blog about bigg data", which in this wek's post discusses the ethical implications of this topic.
65 | --->
66 |
67 | 23. I love this!
68 | --->
69 |
70 | 24. I'm living in Japan, which consistently appears at the better end of the scale.
71 | --->
72 |
73 | 25. In his eloquent reminiscence of his visit to the Kaba- Islam's religious shrine, Mr. Mustafe Akyol uses the word, islamicc Law" several times.
74 | --->
75 |
76 | 26. Fantastic talk.
77 | --->
78 |
79 | 27. This sounds amazing.
80 | --->
81 |
82 | 28. More importantly, there is no government that would legitimize such businesses by issuing business permits or licenses, whose sole purpose is to police their role.
83 | --->
84 |
85 | 29. Aime is an extraordinary speaker.
86 | --->
87 |
88 | 30. I think it's pretty good to see someone stand out and endorse this kind of viewpoint providing a new way to see things.
89 | --->
90 |
91 | 31. I love her voice and the cello.
92 | --->
93 |
94 | 32. He doesn't have a clue about Functional Medicine and its uses in modern Europe.
95 | --->
96 |
97 | 33. Where's the innovation?
98 | --->
99 |
100 | 34. Looking at the data, I was struck not only by Thailand's rapid decrease in fertility rate but also by a similar decrease in fertility rate of its neighbors.
101 | --->
102 |
103 | 35. Many of my friends died in car accidents which could have been all avoided by this technology!
104 | --->
105 |
106 | 36. very nice work mr. klein.
107 | --->
108 |
109 | 37. Muchas veces al ver un problema pensamos que las soluciones siempre son complejas, lo cual implica de cierta forma m´s tiempo y m´s dinero; sin embargo existen soluciones simples y de bajo costo que permiten resultados similares, y eso es lo que vemos en este video que est´ orientado a las clases de escazos recursos.
110 | --->
111 |
112 | 38. the particles he mentions can be descriptions of different possible routs for vibrations of a very complex vortex of interacting stringss".
113 | --->
114 |
115 | 39. Not worthy of Ted.
116 | --->
117 |
118 | 40. not all of us may have the capital and resources that she had but i am pretty sure that most of us have the basic education needed to teach.
119 | --->
120 |
121 | 41. An amazing speaker with knowledge is enough to get me excited about any topic.
122 | --->
123 |
124 | 42. Most of all, it can inspire us to do things we never thought we could.
125 | --->
126 |
127 | 43. I believe that this debate has shown that wind for one can be a big part in solving our energy problems.
128 | --->
129 |
130 | 44. Bjorn reminds me of what I don't really like about organizations and events like Ted, Davos, the Cfr, etc.
131 | --->
132 |
133 | 45. I made the conscience decision to stop listening to music for a while which was devastating because music is my passion by far.
134 | --->
135 |
136 | 46. Many have denied in follow-up that they reject common descent (your 'macroevolution') and claim to have been misled or misrepresented.
137 | --->
138 |
139 | 47. The project requirements were to construct a minimal structure from unusual materials.
140 | --->
141 |
142 | 48. fascinating body of work and process of thinking
143 | --->
144 |
145 | 49. It is a kind of revolution in the area of technology.
146 | --->
147 |
148 | 50. @chris: Have you considered confirmation bias (chery-picking)?
149 | --->
150 |
151 | 51. Are you saying it's impossible for these apes to have a similar consciousness?
152 | --->
153 |
154 | 52. lastt phrase shamelessly copied from who knows who).
155 | --->
156 |
157 | 53. If anything, this lecture should have seen the different type.
158 | --->
159 |
160 | 54. But, its root cause, The Covert Enslavement of Humanity, is history's greatest injustice.
161 | --->
162 |
163 | 55. Dear T Walker sadlyy, you will not understand what that sentence really means".
164 | --->
165 |
166 | 56. Nick Bostrom; you can not prove a negative, self away beings will always search & need meaning, and from what we know from neuroscience, the brain seems to have evolved the ability to experience transcendence and belief.
167 | --->
168 |
169 | 57. This is a bad idea that leads to a generation or more that doesn't understand the fundamentals of calculation.
170 | --->
171 |
172 | 58. A brilliant professor !
173 | --->
174 |
175 | 59. Yet no one has been able to compare the effectiveness of solar orbital microwave power stations?
176 | --->
177 |
178 | 60. As far as his introduction, it was a hook and it was interesting and thought provoking, just like what followed.
179 | --->
180 |
181 | 61. Rahul Dewan: ii feel, the way to prevent terrorism is to 'raise the level of discourse' on God and Spirituality, and make it spread across all religious faiths".
182 | --->
183 |
184 | 62. And equally like the 19th century, our governing powers and ruling elites don't quite have any answer of how to undo what can't really be undone.
185 | --->
186 |
187 | 63. Commenting on Geof Mosher alongg with Tj Green) Iwould start b ystating the fact that evolutionn is war" is quite an understatment.
188 | --->
189 |
190 | 64. I will address them personally if you wish me to do so.
191 | --->
192 |
193 | 65. That's the wonderful right of freedom.
194 | --->
195 |
196 | 66. Very nice.
197 | --->
198 |
199 | 67. But could the opposite be true: that the gregariousness that is the driving force of human beings, is being virtualized by make-believe substitutes to actual social interaction?
200 | --->
201 |
202 | 68. Guys use Vlc player, the mp4 480p version works perfectly on it, awesome quality too.
203 | --->
204 |
205 | 69. Would it be impolite to ask where you stand in this argument?
206 | --->
207 |
208 | 70. Just wow.
209 | --->
210 |
211 | 71. However, had large numbers of Dutch with their hunting and military style rifles been able to start a sniping campaign against the Germans, Wi may have been shortened.
212 | --->
213 |
214 | 72. Amazing - Good Work - He is Worth to not only to India but also to the Whole world His promise for Open Source is also unbelivable a lots of Thanks to him and sponsers
215 | --->
216 |
217 | 73. It isn't good or bad; it's inevitable.
218 | --->
219 |
220 | 74. A couple of years ago the idea was about privacy that if you are connected to the Internet then your privacy is eliminated and you don't have any right to claim it back because you're a part of the network and you're obviously exchanging info that crosses a lot of paths to reach you, but recently, telecom providers are the ones dominating and watching, closely, what everyone is doing.
221 | --->
222 |
223 | 75. Wait for gasoline to hit $5 a gallon and it will.
224 | --->
225 |
226 | 76. Ib is a rigorous international education program for high school students.
227 | --->
228 |
229 | 77. As someone who has spent several years teaching, I can say from experience that children need discipline, a rigid structure and for the most part, they need to be told what to do.
230 | --->
231 |
232 | 78. Horrible architect, decent marketer perhaps?
233 | --->
234 |
235 | 79. They aren't reproducible on command, and they do not openly discussed the incidences because the subject is taboo.
236 | --->
237 |
238 | 80. Examples: To say that conservatives are much more into ideological purity is absurd.
239 | --->
240 |
241 | 81. David is talking about a job interview on steroids, and like all job interviews there is an understanding that things will be shown in the best light possible (to say the least).
242 | --->
243 |
244 | 82. In order to ensure we aren't fluctuations, we need to make an observation and analyze it.
245 | --->
246 |
247 | 83. They don’t have room or numbers and we are losing them.
248 | --->
249 |
250 | 84. The point I'm trying to make is this: a great craftsman is Not by default a great artist.
251 | --->
252 |
253 | 85. They are invited to raise their right hand and in unison say globall Truce Starts Now!
254 | --->
255 |
256 | 86. I would say all the fundamentalisms floating around in peoples minds making war on each other are one of the great problems of our time.
257 | --->
258 |
259 | 87. I will probably be one of the first to be swept away!
260 | --->
261 |
262 | 88. Please don't even Think of taking it down!
263 | --->
264 |
265 | 89. You see a lot of people, youngsters especially, who complain that they are depressed and alone.
266 | --->
267 |
268 | 90. I'm dubious about the payback periods you state.
269 | --->
270 |
271 | 91. Did his disgraceful comment on Israel fall on deaf ears?
272 | --->
273 |
274 | 92. To that end, how would empathy fix any of these problems?
275 | --->
276 |
277 | 93. follow his lead by taking a hard look at your own life and either get help or work on your own to become more 'gay'.
278 | --->
279 |
280 | 94. Sorry, he's a pet peeve.
281 | --->
282 |
283 | 95. human have no right and ability to make it!
284 | --->
285 |
286 | 96. Felix, why should an artist or a political activist not be commercially astute?
287 | --->
288 |
289 | 97. And now we have screwed up our heads were screwing about with monkeys heads as well):
290 | --->
291 |
292 | 98. It searches for and indexes common data in a vast resource and, not surprisingly, ends up finding lots of that data.
293 | --->
294 |
295 | 99. People need to try to be less offended by differing viewpoints.
296 | --->
297 |
298 | 100. their use of memetics suggests an intentionalityy" in memes.
299 | --->
300 |
301 | 101. Noah Feldman constantly refers to the supposed democratic beliefs of a vast majority of Muslims.
302 | --->
303 |
304 | 102. I felt it helped me deal with issues in a good way and was the right attitude to have.
305 | --->
306 |
307 | 103. This feed specific microbes that in tern secrete the nutrient that the plant needs.
308 | --->
309 |
310 | 104. I know that's a bit simplistic, but still.
311 | --->
312 |
313 | 105. another one thing is staffs, they work only for salary they wouldn't give clear leacture.
314 | --->
315 |
316 | 106. Ingenious incorporation of the natural binding aspect of the mycelium from the mushroom plant to create a products)s that could partially compete with some single uses of Styrofoam.
317 | --->
318 |
319 | 107. As for China, I mostly agree with Mr. Tabarok, but I'm not sure if the post-mao development would've been possible without Mao's prior contribution to health seee Hans Rosling's Ted talks).
320 | --->
321 |
322 | 108. his talk is great.
323 | --->
324 |
325 | 109. Now, to briefly address your list of supposed contradictions and fallacies.
326 | --->
327 |
328 | 110. Lol - hilarious.
329 | --->
330 |
331 | 111. But the point is missed what Dan Gilbert is trying to tell you how we proceed looking at some time of value because its illusion, mostly thats what we are buying "that Illusion!
332 | --->
333 |
334 | 112. Stephen Colbert, Jon Stewart and Bill Maher do make things easier to discuss with humor but they talk about American issues and Americans watch them.
335 | --->
336 |
337 | 113. xd It's been awhile since I smiled so much just from a video ^ I want more of this!
338 | --->
339 |
340 | 114. In my view , school is really a good place to sort out the smart ones who survive the education system and create stupid people (to work as labour) .
341 | --->
342 |
343 | 115. Having been a professional background vocalist in Hollywood for several decades, I att age 24) rejoined a pop band, which had drastically changed its repetoire during my absence, having added several vocally demanding rock tunes, which I would be required to sing to regain my old position.
344 | --->
345 |
346 | 116. Thank goodness and for the interactive transcript feature and Lee Smolin.
347 | --->
348 |
349 | 117. I initiated and developed this project some years ago, which makes it possible for disabled people to take a look beyond their horizons and experience the ultimate feeling of freedom and independence: htp:/w.youtube.com/watch?v=ljbmau7jc
350 | --->
351 |
352 | 118. Pinker uses warped statistics ass pointed out in the blog you mention).
353 | --->
354 |
355 | 119. it's also our fault too by allowing men to do so.
356 | --->
357 |
358 | 120. many are institutionalised profit based or religious hagovers from another world ago.
359 | --->
360 |
361 | 121. This is a political manipulation attempt to educate the children.
362 | --->
363 |
364 | 122. It's a glorified 70-year-old hand cannon and it is not modern, civil, humane or precise.
365 | --->
366 |
367 | 123. Peace!
368 | --->
369 |
370 | 124. Of course this may not guarantee.
371 | --->
372 |
373 | 125. This was some kind of parody, right?
374 | --->
375 |
376 | 126. I do these things for a distraction and generally when I am bored.
377 | --->
378 |
379 | 127. right or wrong I think this is an inevitable advancement you have to be fine with it.
380 | --->
381 |
382 | 128. Would really love to go to a postsecret live event.
383 | --->
384 |
385 | 129. " the words from a man, whom I think, seriously meant that.
386 | --->
387 |
388 | 130. I can see that I wasn't as clear in my writing as I thought I was.
389 | --->
390 |
391 | 131. tragic at the least and still playing out every day.
392 | --->
393 |
394 | 132. I wonder how well this was received and how many of his countrymen saw this.
395 | --->
396 |
397 | 133. I am no scientist but it would seem possible to manipulate such large amounts of data on societal reactions and interactions in a way that would produce highly reliable prognostications relative to very specific stimulus.
398 | --->
399 |
400 | 134. While one size will not fit all, this looks to be a wonderful piece in the puzzle.
401 | --->
402 |
403 | 135. any body please help
404 | --->
405 |
406 | 136. This allows our brains to wire for efficiency.
407 | --->
408 |
409 | 137. When I built the house, I built it according to mostt of) the building codes in Canada.
410 | --->
411 |
412 | 138. 2)2 He created them male and female and blessed them and named them [both] Adam [man] at the time they were created.
413 | --->
414 |
415 | 139. I definitely think, that there is no movie, that can be so epic and so truly touching that these real life - stuffs happening here in Ted.
416 | --->
417 |
418 | 140. All reasearch is of course fantastic, and I take your points about satelites seriously, although suspect that most of us could do without sky tv, gps and various military derivatives if we're brutally honest.
419 | --->
420 |
421 |
--------------------------------------------------------------------------------
/TED_comment_annotations/4/complete/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 2043
5 | svn://deepcore.gr/dc/ted/v1/hlt_project/utils/full_corpus/subjects/testing/4/complete
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2013-01-18T18:08:40.681354Z
11 | 1662
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | comment_test.txt
30 | file
31 |
32 |
33 |
34 |
35 | 2013-01-18T19:54:08.000000Z
36 | 302705fd9ecbb863ddb52f325a106d94
37 | 2013-01-18T18:08:40.681354Z
38 | 1662
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 33864
62 |
63 | sentence_test.txt
64 | file
65 |
66 |
67 |
68 |
69 | 2013-01-18T19:54:08.000000Z
70 | be6d1c8983bd2ba64c2c7e29f63ab1ed
71 | 2013-01-18T18:08:40.681354Z
72 | 1662
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 14614
96 |
97 |
--------------------------------------------------------------------------------
/TED_comment_annotations/4/complete/.svn/text-base/sentence_test.txt.svn-base:
--------------------------------------------------------------------------------
1 | 1. No magic, alien assistance or lost mystical powers - just plain patience and determination ignoringg the attendant slavery for the moment).
2 | ---> n
3 |
4 | 2. Not only does he put focus on the dumbing down, dilution and perversion of scientific findings one sees presented in the media, but he explains the mechanisms by which the same findings become inaccessible or incomprehensible to the less probing masses.
5 | ---> n
6 |
7 | 3. This is science at it's best and the future is bright.
8 | ---> p
9 |
10 | 4. Is this really a good idea in the long term?
11 | ---> n
12 |
13 | 5. The inventor did not mean to produce such kind of glue.
14 | ---> n
15 |
16 | 6. People do not want to change now as long as oil is still affordable.
17 | ---> n
18 |
19 | 7. You fail to realize what povertyy" really means in regards to education, connection, status, class, family and social clasism.
20 | ---> n
21 |
22 | 8. clumsy to say the less) but I feel the full conclusion was not reached.
23 | ---> n
24 |
25 | 9. Agricultural techniques may increase that amount but doing so often comes with environmental damage and a loss of biodiversity.
26 | ---> n
27 |
28 | 10. Inspiring talk!
29 | ---> p
30 |
31 | 11. Think what you want, but evolution is true and there's no conspiracy.
32 | ---> n
33 |
34 | 12. The question of great power bringing great responsibility isn't new (spiderman, anyone?
35 | ---> u
36 |
37 | 13. With out ideologues, there is no hope whatsoever.
38 | ---> p
39 |
40 | 14. Beautiful, but for someone like me who's biggest problem is control"l over my experiences, it sounds like a sort of hell to not be able to see my boundaries!
41 | ---> p
42 |
43 | 15. Direction of Magnetic force, and 4.
44 | ---> u
45 |
46 | 16. i got a little bored listening.
47 | ---> n
48 |
49 | 17. Helping Handicapped and/or elderly people.
50 | ---> p
51 |
52 | 18. I do not mean to demean anyone.
53 | ---> n
54 |
55 | 19. I do love this concept of design for outcome, nice talk!
56 | ---> p
57 |
58 | 20. The most basic concept here is people thrive on autonomy.
59 | ---> p
60 |
61 | 21. The negative attitude is ingrained in us by evolution, no less.
62 | ---> n
63 |
64 | 22. Check out this newly created blog about bigg data", which in this wek's post discusses the ethical implications of this topic.
65 | ---> p
66 |
67 | 23. I love this!
68 | ---> p
69 |
70 | 24. I'm living in Japan, which consistently appears at the better end of the scale.
71 | ---> p
72 |
73 | 25. In his eloquent reminiscence of his visit to the Kaba- Islam's religious shrine, Mr. Mustafe Akyol uses the word, islamicc Law" several times.
74 | ---> u
75 |
76 | 26. Fantastic talk.
77 | ---> p
78 |
79 | 27. This sounds amazing.
80 | ---> p
81 |
82 | 28. More importantly, there is no government that would legitimize such businesses by issuing business permits or licenses, whose sole purpose is to police their role.
83 | ---> n
84 |
85 | 29. Aime is an extraordinary speaker.
86 | ---> p
87 |
88 | 30. I think it's pretty good to see someone stand out and endorse this kind of viewpoint providing a new way to see things.
89 | ---> p
90 |
91 | 31. I love her voice and the cello.
92 | ---> p
93 |
94 | 32. He doesn't have a clue about Functional Medicine and its uses in modern Europe.
95 | ---> n
96 |
97 | 33. Where's the innovation?
98 | ---> n
99 |
100 | 34. Looking at the data, I was struck not only by Thailand's rapid decrease in fertility rate but also by a similar decrease in fertility rate of its neighbors.
101 | ---> n
102 |
103 | 35. Many of my friends died in car accidents which could have been all avoided by this technology!
104 | ---> p
105 |
106 | 36. very nice work mr. klein.
107 | ---> p
108 |
109 | 37. Muchas veces al ver un problema pensamos que las soluciones siempre son complejas, lo cual implica de cierta forma m´s tiempo y m´s dinero; sin embargo existen soluciones simples y de bajo costo que permiten resultados similares, y eso es lo que vemos en este video que est´ orientado a las clases de escazos recursos.
110 | ---> p
111 |
112 | 38. the particles he mentions can be descriptions of different possible routs for vibrations of a very complex vortex of interacting stringss".
113 | ---> p
114 |
115 | 39. Not worthy of Ted.
116 | ---> n
117 |
118 | 40. not all of us may have the capital and resources that she had but i am pretty sure that most of us have the basic education needed to teach.
119 | ---> p
120 |
121 | 41. An amazing speaker with knowledge is enough to get me excited about any topic.
122 | ---> p
123 |
124 | 42. Most of all, it can inspire us to do things we never thought we could.
125 | ---> p
126 |
127 | 43. I believe that this debate has shown that wind for one can be a big part in solving our energy problems.
128 | ---> p
129 |
130 | 44. Bjorn reminds me of what I don't really like about organizations and events like Ted, Davos, the Cfr, etc.
131 | ---> n
132 |
133 | 45. I made the conscience decision to stop listening to music for a while which was devastating because music is my passion by far.
134 | ---> p
135 |
136 | 46. Many have denied in follow-up that they reject common descent (your 'macroevolution') and claim to have been misled or misrepresented.
137 | ---> n
138 |
139 | 47. The project requirements were to construct a minimal structure from unusual materials.
140 | ---> n
141 |
142 | 48. fascinating body of work and process of thinking
143 | ---> p
144 |
145 | 49. It is a kind of revolution in the area of technology.
146 | ---> p
147 |
148 | 50. @chris: Have you considered confirmation bias (chery-picking)?
149 | ---> n
150 |
151 | 51. Are you saying it's impossible for these apes to have a similar consciousness?
152 | ---> n
153 |
154 | 52. lastt phrase shamelessly copied from who knows who).
155 | ---> n
156 |
157 | 53. If anything, this lecture should have seen the different type.
158 | ---> n
159 |
160 | 54. But, its root cause, The Covert Enslavement of Humanity, is history's greatest injustice.
161 | ---> n
162 |
163 | 55. Dear T Walker sadlyy, you will not understand what that sentence really means".
164 | ---> n
165 |
166 | 56. Nick Bostrom; you can not prove a negative, self away beings will always search & need meaning, and from what we know from neuroscience, the brain seems to have evolved the ability to experience transcendence and belief.
167 | ---> n
168 |
169 | 57. This is a bad idea that leads to a generation or more that doesn't understand the fundamentals of calculation.
170 | ---> n
171 |
172 | 58. A brilliant professor !
173 | ---> p
174 |
175 | 59. Yet no one has been able to compare the effectiveness of solar orbital microwave power stations?
176 | ---> n
177 |
178 | 60. As far as his introduction, it was a hook and it was interesting and thought provoking, just like what followed.
179 | ---> p
180 |
181 | 61. Rahul Dewan: ii feel, the way to prevent terrorism is to 'raise the level of discourse' on God and Spirituality, and make it spread across all religious faiths".
182 | ---> p
183 |
184 | 62. And equally like the 19th century, our governing powers and ruling elites don't quite have any answer of how to undo what can't really be undone.
185 | ---> n
186 |
187 | 63. Commenting on Geof Mosher alongg with Tj Green) Iwould start b ystating the fact that evolutionn is war" is quite an understatment.
188 | ---> n
189 |
190 | 64. I will address them personally if you wish me to do so.
191 | ---> n
192 |
193 | 65. That's the wonderful right of freedom.
194 | ---> p
195 |
196 | 66. Very nice.
197 | ---> p
198 |
199 | 67. But could the opposite be true: that the gregariousness that is the driving force of human beings, is being virtualized by make-believe substitutes to actual social interaction?
200 | ---> p
201 |
202 | 68. Guys use Vlc player, the mp4 480p version works perfectly on it, awesome quality too.
203 | ---> u
204 |
205 | 69. Would it be impolite to ask where you stand in this argument?
206 | ---> n
207 |
208 | 70. Just wow.
209 | ---> p
210 |
211 | 71. However, had large numbers of Dutch with their hunting and military style rifles been able to start a sniping campaign against the Germans, Wi may have been shortened.
212 | ---> n
213 |
214 | 72. Amazing - Good Work - He is Worth to not only to India but also to the Whole world His promise for Open Source is also unbelivable a lots of Thanks to him and sponsers
215 | ---> p
216 |
217 | 73. It isn't good or bad; it's inevitable.
218 | ---> u
219 |
220 | 74. A couple of years ago the idea was about privacy that if you are connected to the Internet then your privacy is eliminated and you don't have any right to claim it back because you're a part of the network and you're obviously exchanging info that crosses a lot of paths to reach you, but recently, telecom providers are the ones dominating and watching, closely, what everyone is doing.
221 | ---> n
222 |
223 | 75. Wait for gasoline to hit $5 a gallon and it will.
224 | ---> n
225 |
226 | 76. Ib is a rigorous international education program for high school students.
227 | ---> p
228 |
229 | 77. As someone who has spent several years teaching, I can say from experience that children need discipline, a rigid structure and for the most part, they need to be told what to do.
230 | ---> n
231 |
232 | 78. Horrible architect, decent marketer perhaps?
233 | ---> n
234 |
235 | 79. They aren't reproducible on command, and they do not openly discussed the incidences because the subject is taboo.
236 | ---> n
237 |
238 | 80. Examples: To say that conservatives are much more into ideological purity is absurd.
239 | ---> n
240 |
241 | 81. David is talking about a job interview on steroids, and like all job interviews there is an understanding that things will be shown in the best light possible (to say the least).
242 | ---> p
243 |
244 | 82. In order to ensure we aren't fluctuations, we need to make an observation and analyze it.
245 | ---> n
246 |
247 | 83. They don’t have room or numbers and we are losing them.
248 | ---> n
249 |
250 | 84. The point I'm trying to make is this: a great craftsman is Not by default a great artist.
251 | ---> n
252 |
253 | 85. They are invited to raise their right hand and in unison say globall Truce Starts Now!
254 | ---> u
255 |
256 | 86. I would say all the fundamentalisms floating around in peoples minds making war on each other are one of the great problems of our time.
257 | ---> u
258 |
259 | 87. I will probably be one of the first to be swept away!
260 | ---> p
261 |
262 | 88. Please don't even Think of taking it down!
263 | ---> u
264 |
265 | 89. You see a lot of people, youngsters especially, who complain that they are depressed and alone.
266 | ---> p
267 |
268 | 90. I'm dubious about the payback periods you state.
269 | ---> n
270 |
271 | 91. Did his disgraceful comment on Israel fall on deaf ears?
272 | ---> n
273 |
274 | 92. To that end, how would empathy fix any of these problems?
275 | ---> n
276 |
277 | 93. follow his lead by taking a hard look at your own life and either get help or work on your own to become more 'gay'.
278 | ---> n
279 |
280 | 94. Sorry, he's a pet peeve.
281 | ---> n
282 |
283 | 95. human have no right and ability to make it!
284 | ---> n
285 |
286 | 96. Felix, why should an artist or a political activist not be commercially astute?
287 | ---> n
288 |
289 | 97. And now we have screwed up our heads were screwing about with monkeys heads as well):
290 | ---> n
291 |
292 | 98. It searches for and indexes common data in a vast resource and, not surprisingly, ends up finding lots of that data.
293 | ---> n
294 |
295 | 99. People need to try to be less offended by differing viewpoints.
296 | ---> p
297 |
298 | 100. their use of memetics suggests an intentionalityy" in memes.
299 | ---> u
300 |
301 | 101. Noah Feldman constantly refers to the supposed democratic beliefs of a vast majority of Muslims.
302 | ---> n
303 |
304 | 102. I felt it helped me deal with issues in a good way and was the right attitude to have.
305 | ---> p
306 |
307 | 103. This feed specific microbes that in tern secrete the nutrient that the plant needs.
308 | ---> p
309 |
310 | 104. I know that's a bit simplistic, but still.
311 | ---> u
312 |
313 | 105. another one thing is staffs, they work only for salary they wouldn't give clear leacture.
314 | ---> n
315 |
316 | 106. Ingenious incorporation of the natural binding aspect of the mycelium from the mushroom plant to create a products)s that could partially compete with some single uses of Styrofoam.
317 | ---> p
318 |
319 | 107. As for China, I mostly agree with Mr. Tabarok, but I'm not sure if the post-mao development would've been possible without Mao's prior contribution to health seee Hans Rosling's Ted talks).
320 | ---> p
321 |
322 | 108. his talk is great.
323 | ---> p
324 |
325 | 109. Now, to briefly address your list of supposed contradictions and fallacies.
326 | ---> n
327 |
328 | 110. Lol - hilarious.
329 | ---> p
330 |
331 | 111. But the point is missed what Dan Gilbert is trying to tell you how we proceed looking at some time of value because its illusion, mostly thats what we are buying "that Illusion!
332 | ---> p
333 |
334 | 112. Stephen Colbert, Jon Stewart and Bill Maher do make things easier to discuss with humor but they talk about American issues and Americans watch them.
335 | ---> n
336 |
337 | 113. xd It's been awhile since I smiled so much just from a video ^ I want more of this!
338 | ---> p
339 |
340 | 114. In my view , school is really a good place to sort out the smart ones who survive the education system and create stupid people (to work as labour) .
341 | ---> u
342 |
343 | 115. Having been a professional background vocalist in Hollywood for several decades, I att age 24) rejoined a pop band, which had drastically changed its repetoire during my absence, having added several vocally demanding rock tunes, which I would be required to sing to regain my old position.
344 | ---> n
345 |
346 | 116. Thank goodness and for the interactive transcript feature and Lee Smolin.
347 | ---> p
348 |
349 | 117. I initiated and developed this project some years ago, which makes it possible for disabled people to take a look beyond their horizons and experience the ultimate feeling of freedom and independence: htp:/w.youtube.com/watch?v=ljbmau7jc
350 | ---> p
351 |
352 | 118. Pinker uses warped statistics ass pointed out in the blog you mention).
353 | ---> n
354 |
355 | 119. it's also our fault too by allowing men to do so.
356 | ---> n
357 |
358 | 120. many are institutionalised profit based or religious hagovers from another world ago.
359 | ---> p
360 |
361 | 121. This is a political manipulation attempt to educate the children.
362 | ---> n
363 |
364 | 122. It's a glorified 70-year-old hand cannon and it is not modern, civil, humane or precise.
365 | ---> n
366 |
367 | 123. Peace!
368 | ---> p
369 |
370 | 124. Of course this may not guarantee.
371 | ---> n
372 |
373 | 125. This was some kind of parody, right?
374 | ---> n
375 |
376 | 126. I do these things for a distraction and generally when I am bored.
377 | ---> n
378 |
379 | 127. right or wrong I think this is an inevitable advancement you have to be fine with it.
380 | ---> p
381 |
382 | 128. Would really love to go to a postsecret live event.
383 | ---> p
384 |
385 | 129. " the words from a man, whom I think, seriously meant that.
386 | ---> n
387 |
388 | 130. I can see that I wasn't as clear in my writing as I thought I was.
389 | ---> u
390 |
391 | 131. tragic at the least and still playing out every day.
392 | ---> n
393 |
394 | 132. I wonder how well this was received and how many of his countrymen saw this.
395 | ---> n
396 |
397 | 133. I am no scientist but it would seem possible to manipulate such large amounts of data on societal reactions and interactions in a way that would produce highly reliable prognostications relative to very specific stimulus.
398 | ---> p
399 |
400 | 134. While one size will not fit all, this looks to be a wonderful piece in the puzzle.
401 | ---> p
402 |
403 | 135. any body please help
404 | ---> n
405 |
406 | 136. This allows our brains to wire for efficiency.
407 | ---> p
408 |
409 | 137. When I built the house, I built it according to mostt of) the building codes in Canada.
410 | ---> u
411 |
412 | 138. 2)2 He created them male and female and blessed them and named them [both] Adam [man] at the time they were created.
413 | ---> u
414 |
415 | 139. I definitely think, that there is no movie, that can be so epic and so truly touching that these real life - stuffs happening here in Ted.
416 | ---> p
417 |
418 | 140. All reasearch is of course fantastic, and I take your points about satelites seriously, although suspect that most of us could do without sky tv, gps and various military derivatives if we're brutally honest.
419 | ---> n
420 |
421 |
--------------------------------------------------------------------------------
/TED_comment_annotations/4/complete/sentence_test.txt:
--------------------------------------------------------------------------------
1 | 1. No magic, alien assistance or lost mystical powers - just plain patience and determination ignoringg the attendant slavery for the moment).
2 | ---> n
3 |
4 | 2. Not only does he put focus on the dumbing down, dilution and perversion of scientific findings one sees presented in the media, but he explains the mechanisms by which the same findings become inaccessible or incomprehensible to the less probing masses.
5 | ---> n
6 |
7 | 3. This is science at it's best and the future is bright.
8 | ---> p
9 |
10 | 4. Is this really a good idea in the long term?
11 | ---> n
12 |
13 | 5. The inventor did not mean to produce such kind of glue.
14 | ---> n
15 |
16 | 6. People do not want to change now as long as oil is still affordable.
17 | ---> n
18 |
19 | 7. You fail to realize what povertyy" really means in regards to education, connection, status, class, family and social clasism.
20 | ---> n
21 |
22 | 8. clumsy to say the less) but I feel the full conclusion was not reached.
23 | ---> n
24 |
25 | 9. Agricultural techniques may increase that amount but doing so often comes with environmental damage and a loss of biodiversity.
26 | ---> n
27 |
28 | 10. Inspiring talk!
29 | ---> p
30 |
31 | 11. Think what you want, but evolution is true and there's no conspiracy.
32 | ---> n
33 |
34 | 12. The question of great power bringing great responsibility isn't new (spiderman, anyone?
35 | ---> u
36 |
37 | 13. With out ideologues, there is no hope whatsoever.
38 | ---> p
39 |
40 | 14. Beautiful, but for someone like me who's biggest problem is control"l over my experiences, it sounds like a sort of hell to not be able to see my boundaries!
41 | ---> p
42 |
43 | 15. Direction of Magnetic force, and 4.
44 | ---> u
45 |
46 | 16. i got a little bored listening.
47 | ---> n
48 |
49 | 17. Helping Handicapped and/or elderly people.
50 | ---> p
51 |
52 | 18. I do not mean to demean anyone.
53 | ---> n
54 |
55 | 19. I do love this concept of design for outcome, nice talk!
56 | ---> p
57 |
58 | 20. The most basic concept here is people thrive on autonomy.
59 | ---> p
60 |
61 | 21. The negative attitude is ingrained in us by evolution, no less.
62 | ---> n
63 |
64 | 22. Check out this newly created blog about bigg data", which in this wek's post discusses the ethical implications of this topic.
65 | ---> p
66 |
67 | 23. I love this!
68 | ---> p
69 |
70 | 24. I'm living in Japan, which consistently appears at the better end of the scale.
71 | ---> p
72 |
73 | 25. In his eloquent reminiscence of his visit to the Kaba- Islam's religious shrine, Mr. Mustafe Akyol uses the word, islamicc Law" several times.
74 | ---> u
75 |
76 | 26. Fantastic talk.
77 | ---> p
78 |
79 | 27. This sounds amazing.
80 | ---> p
81 |
82 | 28. More importantly, there is no government that would legitimize such businesses by issuing business permits or licenses, whose sole purpose is to police their role.
83 | ---> n
84 |
85 | 29. Aime is an extraordinary speaker.
86 | ---> p
87 |
88 | 30. I think it's pretty good to see someone stand out and endorse this kind of viewpoint providing a new way to see things.
89 | ---> p
90 |
91 | 31. I love her voice and the cello.
92 | ---> p
93 |
94 | 32. He doesn't have a clue about Functional Medicine and its uses in modern Europe.
95 | ---> n
96 |
97 | 33. Where's the innovation?
98 | ---> n
99 |
100 | 34. Looking at the data, I was struck not only by Thailand's rapid decrease in fertility rate but also by a similar decrease in fertility rate of its neighbors.
101 | ---> n
102 |
103 | 35. Many of my friends died in car accidents which could have been all avoided by this technology!
104 | ---> p
105 |
106 | 36. very nice work mr. klein.
107 | ---> p
108 |
109 | 37. Muchas veces al ver un problema pensamos que las soluciones siempre son complejas, lo cual implica de cierta forma m´s tiempo y m´s dinero; sin embargo existen soluciones simples y de bajo costo que permiten resultados similares, y eso es lo que vemos en este video que est´ orientado a las clases de escazos recursos.
110 | ---> p
111 |
112 | 38. the particles he mentions can be descriptions of different possible routs for vibrations of a very complex vortex of interacting stringss".
113 | ---> p
114 |
115 | 39. Not worthy of Ted.
116 | ---> n
117 |
118 | 40. not all of us may have the capital and resources that she had but i am pretty sure that most of us have the basic education needed to teach.
119 | ---> p
120 |
121 | 41. An amazing speaker with knowledge is enough to get me excited about any topic.
122 | ---> p
123 |
124 | 42. Most of all, it can inspire us to do things we never thought we could.
125 | ---> p
126 |
127 | 43. I believe that this debate has shown that wind for one can be a big part in solving our energy problems.
128 | ---> p
129 |
130 | 44. Bjorn reminds me of what I don't really like about organizations and events like Ted, Davos, the Cfr, etc.
131 | ---> n
132 |
133 | 45. I made the conscience decision to stop listening to music for a while which was devastating because music is my passion by far.
134 | ---> p
135 |
136 | 46. Many have denied in follow-up that they reject common descent (your 'macroevolution') and claim to have been misled or misrepresented.
137 | ---> n
138 |
139 | 47. The project requirements were to construct a minimal structure from unusual materials.
140 | ---> n
141 |
142 | 48. fascinating body of work and process of thinking
143 | ---> p
144 |
145 | 49. It is a kind of revolution in the area of technology.
146 | ---> p
147 |
148 | 50. @chris: Have you considered confirmation bias (chery-picking)?
149 | ---> n
150 |
151 | 51. Are you saying it's impossible for these apes to have a similar consciousness?
152 | ---> n
153 |
154 | 52. lastt phrase shamelessly copied from who knows who).
155 | ---> n
156 |
157 | 53. If anything, this lecture should have seen the different type.
158 | ---> n
159 |
160 | 54. But, its root cause, The Covert Enslavement of Humanity, is history's greatest injustice.
161 | ---> n
162 |
163 | 55. Dear T Walker sadlyy, you will not understand what that sentence really means".
164 | ---> n
165 |
166 | 56. Nick Bostrom; you can not prove a negative, self away beings will always search & need meaning, and from what we know from neuroscience, the brain seems to have evolved the ability to experience transcendence and belief.
167 | ---> n
168 |
169 | 57. This is a bad idea that leads to a generation or more that doesn't understand the fundamentals of calculation.
170 | ---> n
171 |
172 | 58. A brilliant professor !
173 | ---> p
174 |
175 | 59. Yet no one has been able to compare the effectiveness of solar orbital microwave power stations?
176 | ---> n
177 |
178 | 60. As far as his introduction, it was a hook and it was interesting and thought provoking, just like what followed.
179 | ---> p
180 |
181 | 61. Rahul Dewan: ii feel, the way to prevent terrorism is to 'raise the level of discourse' on God and Spirituality, and make it spread across all religious faiths".
182 | ---> p
183 |
184 | 62. And equally like the 19th century, our governing powers and ruling elites don't quite have any answer of how to undo what can't really be undone.
185 | ---> n
186 |
187 | 63. Commenting on Geof Mosher alongg with Tj Green) Iwould start b ystating the fact that evolutionn is war" is quite an understatment.
188 | ---> n
189 |
190 | 64. I will address them personally if you wish me to do so.
191 | ---> n
192 |
193 | 65. That's the wonderful right of freedom.
194 | ---> p
195 |
196 | 66. Very nice.
197 | ---> p
198 |
199 | 67. But could the opposite be true: that the gregariousness that is the driving force of human beings, is being virtualized by make-believe substitutes to actual social interaction?
200 | ---> p
201 |
202 | 68. Guys use Vlc player, the mp4 480p version works perfectly on it, awesome quality too.
203 | ---> u
204 |
205 | 69. Would it be impolite to ask where you stand in this argument?
206 | ---> n
207 |
208 | 70. Just wow.
209 | ---> p
210 |
211 | 71. However, had large numbers of Dutch with their hunting and military style rifles been able to start a sniping campaign against the Germans, Wi may have been shortened.
212 | ---> n
213 |
214 | 72. Amazing - Good Work - He is Worth to not only to India but also to the Whole world His promise for Open Source is also unbelivable a lots of Thanks to him and sponsers
215 | ---> p
216 |
217 | 73. It isn't good or bad; it's inevitable.
218 | ---> u
219 |
220 | 74. A couple of years ago the idea was about privacy that if you are connected to the Internet then your privacy is eliminated and you don't have any right to claim it back because you're a part of the network and you're obviously exchanging info that crosses a lot of paths to reach you, but recently, telecom providers are the ones dominating and watching, closely, what everyone is doing.
221 | ---> n
222 |
223 | 75. Wait for gasoline to hit $5 a gallon and it will.
224 | ---> n
225 |
226 | 76. Ib is a rigorous international education program for high school students.
227 | ---> p
228 |
229 | 77. As someone who has spent several years teaching, I can say from experience that children need discipline, a rigid structure and for the most part, they need to be told what to do.
230 | ---> n
231 |
232 | 78. Horrible architect, decent marketer perhaps?
233 | ---> n
234 |
235 | 79. They aren't reproducible on command, and they do not openly discussed the incidences because the subject is taboo.
236 | ---> n
237 |
238 | 80. Examples: To say that conservatives are much more into ideological purity is absurd.
239 | ---> n
240 |
241 | 81. David is talking about a job interview on steroids, and like all job interviews there is an understanding that things will be shown in the best light possible (to say the least).
242 | ---> p
243 |
244 | 82. In order to ensure we aren't fluctuations, we need to make an observation and analyze it.
245 | ---> n
246 |
247 | 83. They don’t have room or numbers and we are losing them.
248 | ---> n
249 |
250 | 84. The point I'm trying to make is this: a great craftsman is Not by default a great artist.
251 | ---> n
252 |
253 | 85. They are invited to raise their right hand and in unison say globall Truce Starts Now!
254 | ---> u
255 |
256 | 86. I would say all the fundamentalisms floating around in peoples minds making war on each other are one of the great problems of our time.
257 | ---> u
258 |
259 | 87. I will probably be one of the first to be swept away!
260 | ---> p
261 |
262 | 88. Please don't even Think of taking it down!
263 | ---> u
264 |
265 | 89. You see a lot of people, youngsters especially, who complain that they are depressed and alone.
266 | ---> p
267 |
268 | 90. I'm dubious about the payback periods you state.
269 | ---> n
270 |
271 | 91. Did his disgraceful comment on Israel fall on deaf ears?
272 | ---> n
273 |
274 | 92. To that end, how would empathy fix any of these problems?
275 | ---> n
276 |
277 | 93. follow his lead by taking a hard look at your own life and either get help or work on your own to become more 'gay'.
278 | ---> n
279 |
280 | 94. Sorry, he's a pet peeve.
281 | ---> n
282 |
283 | 95. human have no right and ability to make it!
284 | ---> n
285 |
286 | 96. Felix, why should an artist or a political activist not be commercially astute?
287 | ---> n
288 |
289 | 97. And now we have screwed up our heads were screwing about with monkeys heads as well):
290 | ---> n
291 |
292 | 98. It searches for and indexes common data in a vast resource and, not surprisingly, ends up finding lots of that data.
293 | ---> n
294 |
295 | 99. People need to try to be less offended by differing viewpoints.
296 | ---> p
297 |
298 | 100. their use of memetics suggests an intentionalityy" in memes.
299 | ---> u
300 |
301 | 101. Noah Feldman constantly refers to the supposed democratic beliefs of a vast majority of Muslims.
302 | ---> n
303 |
304 | 102. I felt it helped me deal with issues in a good way and was the right attitude to have.
305 | ---> p
306 |
307 | 103. This feed specific microbes that in tern secrete the nutrient that the plant needs.
308 | ---> p
309 |
310 | 104. I know that's a bit simplistic, but still.
311 | ---> u
312 |
313 | 105. another one thing is staffs, they work only for salary they wouldn't give clear leacture.
314 | ---> n
315 |
316 | 106. Ingenious incorporation of the natural binding aspect of the mycelium from the mushroom plant to create a products)s that could partially compete with some single uses of Styrofoam.
317 | ---> p
318 |
319 | 107. As for China, I mostly agree with Mr. Tabarok, but I'm not sure if the post-mao development would've been possible without Mao's prior contribution to health seee Hans Rosling's Ted talks).
320 | ---> p
321 |
322 | 108. his talk is great.
323 | ---> p
324 |
325 | 109. Now, to briefly address your list of supposed contradictions and fallacies.
326 | ---> n
327 |
328 | 110. Lol - hilarious.
329 | ---> p
330 |
331 | 111. But the point is missed what Dan Gilbert is trying to tell you how we proceed looking at some time of value because its illusion, mostly thats what we are buying "that Illusion!
332 | ---> p
333 |
334 | 112. Stephen Colbert, Jon Stewart and Bill Maher do make things easier to discuss with humor but they talk about American issues and Americans watch them.
335 | ---> n
336 |
337 | 113. xd It's been awhile since I smiled so much just from a video ^ I want more of this!
338 | ---> p
339 |
340 | 114. In my view , school is really a good place to sort out the smart ones who survive the education system and create stupid people (to work as labour) .
341 | ---> u
342 |
343 | 115. Having been a professional background vocalist in Hollywood for several decades, I att age 24) rejoined a pop band, which had drastically changed its repetoire during my absence, having added several vocally demanding rock tunes, which I would be required to sing to regain my old position.
344 | ---> n
345 |
346 | 116. Thank goodness and for the interactive transcript feature and Lee Smolin.
347 | ---> p
348 |
349 | 117. I initiated and developed this project some years ago, which makes it possible for disabled people to take a look beyond their horizons and experience the ultimate feeling of freedom and independence: htp:/w.youtube.com/watch?v=ljbmau7jc
350 | ---> p
351 |
352 | 118. Pinker uses warped statistics ass pointed out in the blog you mention).
353 | ---> n
354 |
355 | 119. it's also our fault too by allowing men to do so.
356 | ---> n
357 |
358 | 120. many are institutionalised profit based or religious hagovers from another world ago.
359 | ---> p
360 |
361 | 121. This is a political manipulation attempt to educate the children.
362 | ---> n
363 |
364 | 122. It's a glorified 70-year-old hand cannon and it is not modern, civil, humane or precise.
365 | ---> n
366 |
367 | 123. Peace!
368 | ---> p
369 |
370 | 124. Of course this may not guarantee.
371 | ---> n
372 |
373 | 125. This was some kind of parody, right?
374 | ---> n
375 |
376 | 126. I do these things for a distraction and generally when I am bored.
377 | ---> n
378 |
379 | 127. right or wrong I think this is an inevitable advancement you have to be fine with it.
380 | ---> p
381 |
382 | 128. Would really love to go to a postsecret live event.
383 | ---> p
384 |
385 | 129. " the words from a man, whom I think, seriously meant that.
386 | ---> n
387 |
388 | 130. I can see that I wasn't as clear in my writing as I thought I was.
389 | ---> u
390 |
391 | 131. tragic at the least and still playing out every day.
392 | ---> n
393 |
394 | 132. I wonder how well this was received and how many of his countrymen saw this.
395 | ---> n
396 |
397 | 133. I am no scientist but it would seem possible to manipulate such large amounts of data on societal reactions and interactions in a way that would produce highly reliable prognostications relative to very specific stimulus.
398 | ---> p
399 |
400 | 134. While one size will not fit all, this looks to be a wonderful piece in the puzzle.
401 | ---> p
402 |
403 | 135. any body please help
404 | ---> n
405 |
406 | 136. This allows our brains to wire for efficiency.
407 | ---> p
408 |
409 | 137. When I built the house, I built it according to mostt of) the building codes in Canada.
410 | ---> u
411 |
412 | 138. 2)2 He created them male and female and blessed them and named them [both] Adam [man] at the time they were created.
413 | ---> u
414 |
415 | 139. I definitely think, that there is no movie, that can be so epic and so truly touching that these real life - stuffs happening here in Ted.
416 | ---> p
417 |
418 | 140. All reasearch is of course fantastic, and I take your points about satelites seriously, although suspect that most of us could do without sky tv, gps and various military derivatives if we're brutally honest.
419 | ---> n
420 |
421 |
--------------------------------------------------------------------------------
/TED_comment_annotations/4/sentence_test.txt:
--------------------------------------------------------------------------------
1 | 1. No magic, alien assistance or lost mystical powers - just plain patience and determination ignoringg the attendant slavery for the moment).
2 | --->
3 |
4 | 2. Not only does he put focus on the dumbing down, dilution and perversion of scientific findings one sees presented in the media, but he explains the mechanisms by which the same findings become inaccessible or incomprehensible to the less probing masses.
5 | --->
6 |
7 | 3. This is science at it's best and the future is bright.
8 | --->
9 |
10 | 4. Is this really a good idea in the long term?
11 | --->
12 |
13 | 5. The inventor did not mean to produce such kind of glue.
14 | --->
15 |
16 | 6. People do not want to change now as long as oil is still affordable.
17 | --->
18 |
19 | 7. You fail to realize what povertyy" really means in regards to education, connection, status, class, family and social clasism.
20 | --->
21 |
22 | 8. clumsy to say the less) but I feel the full conclusion was not reached.
23 | --->
24 |
25 | 9. Agricultural techniques may increase that amount but doing so often comes with environmental damage and a loss of biodiversity.
26 | --->
27 |
28 | 10. Inspiring talk!
29 | --->
30 |
31 | 11. Think what you want, but evolution is true and there's no conspiracy.
32 | --->
33 |
34 | 12. The question of great power bringing great responsibility isn't new (spiderman, anyone?
35 | --->
36 |
37 | 13. With out ideologues, there is no hope whatsoever.
38 | --->
39 |
40 | 14. Beautiful, but for someone like me who's biggest problem is control"l over my experiences, it sounds like a sort of hell to not be able to see my boundaries!
41 | --->
42 |
43 | 15. Direction of Magnetic force, and 4.
44 | --->
45 |
46 | 16. i got a little bored listening.
47 | --->
48 |
49 | 17. Helping Handicapped and/or elderly people.
50 | --->
51 |
52 | 18. I do not mean to demean anyone.
53 | --->
54 |
55 | 19. I do love this concept of design for outcome, nice talk!
56 | --->
57 |
58 | 20. The most basic concept here is people thrive on autonomy.
59 | --->
60 |
61 | 21. The negative attitude is ingrained in us by evolution, no less.
62 | --->
63 |
64 | 22. Check out this newly created blog about bigg data", which in this wek's post discusses the ethical implications of this topic.
65 | --->
66 |
67 | 23. I love this!
68 | --->
69 |
70 | 24. I'm living in Japan, which consistently appears at the better end of the scale.
71 | --->
72 |
73 | 25. In his eloquent reminiscence of his visit to the Kaba- Islam's religious shrine, Mr. Mustafe Akyol uses the word, islamicc Law" several times.
74 | --->
75 |
76 | 26. Fantastic talk.
77 | --->
78 |
79 | 27. This sounds amazing.
80 | --->
81 |
82 | 28. More importantly, there is no government that would legitimize such businesses by issuing business permits or licenses, whose sole purpose is to police their role.
83 | --->
84 |
85 | 29. Aime is an extraordinary speaker.
86 | --->
87 |
88 | 30. I think it's pretty good to see someone stand out and endorse this kind of viewpoint providing a new way to see things.
89 | --->
90 |
91 | 31. I love her voice and the cello.
92 | --->
93 |
94 | 32. He doesn't have a clue about Functional Medicine and its uses in modern Europe.
95 | --->
96 |
97 | 33. Where's the innovation?
98 | --->
99 |
100 | 34. Looking at the data, I was struck not only by Thailand's rapid decrease in fertility rate but also by a similar decrease in fertility rate of its neighbors.
101 | --->
102 |
103 | 35. Many of my friends died in car accidents which could have been all avoided by this technology!
104 | --->
105 |
106 | 36. very nice work mr. klein.
107 | --->
108 |
109 | 37. Muchas veces al ver un problema pensamos que las soluciones siempre son complejas, lo cual implica de cierta forma m´s tiempo y m´s dinero; sin embargo existen soluciones simples y de bajo costo que permiten resultados similares, y eso es lo que vemos en este video que est´ orientado a las clases de escazos recursos.
110 | --->
111 |
112 | 38. the particles he mentions can be descriptions of different possible routs for vibrations of a very complex vortex of interacting stringss".
113 | --->
114 |
115 | 39. Not worthy of Ted.
116 | --->
117 |
118 | 40. not all of us may have the capital and resources that she had but i am pretty sure that most of us have the basic education needed to teach.
119 | --->
120 |
121 | 41. An amazing speaker with knowledge is enough to get me excited about any topic.
122 | --->
123 |
124 | 42. Most of all, it can inspire us to do things we never thought we could.
125 | --->
126 |
127 | 43. I believe that this debate has shown that wind for one can be a big part in solving our energy problems.
128 | --->
129 |
130 | 44. Bjorn reminds me of what I don't really like about organizations and events like Ted, Davos, the Cfr, etc.
131 | --->
132 |
133 | 45. I made the conscience decision to stop listening to music for a while which was devastating because music is my passion by far.
134 | --->
135 |
136 | 46. Many have denied in follow-up that they reject common descent (your 'macroevolution') and claim to have been misled or misrepresented.
137 | --->
138 |
139 | 47. The project requirements were to construct a minimal structure from unusual materials.
140 | --->
141 |
142 | 48. fascinating body of work and process of thinking
143 | --->
144 |
145 | 49. It is a kind of revolution in the area of technology.
146 | --->
147 |
148 | 50. @chris: Have you considered confirmation bias (chery-picking)?
149 | --->
150 |
151 | 51. Are you saying it's impossible for these apes to have a similar consciousness?
152 | --->
153 |
154 | 52. lastt phrase shamelessly copied from who knows who).
155 | --->
156 |
157 | 53. If anything, this lecture should have seen the different type.
158 | --->
159 |
160 | 54. But, its root cause, The Covert Enslavement of Humanity, is history's greatest injustice.
161 | --->
162 |
163 | 55. Dear T Walker sadlyy, you will not understand what that sentence really means".
164 | --->
165 |
166 | 56. Nick Bostrom; you can not prove a negative, self away beings will always search & need meaning, and from what we know from neuroscience, the brain seems to have evolved the ability to experience transcendence and belief.
167 | --->
168 |
169 | 57. This is a bad idea that leads to a generation or more that doesn't understand the fundamentals of calculation.
170 | --->
171 |
172 | 58. A brilliant professor !
173 | --->
174 |
175 | 59. Yet no one has been able to compare the effectiveness of solar orbital microwave power stations?
176 | --->
177 |
178 | 60. As far as his introduction, it was a hook and it was interesting and thought provoking, just like what followed.
179 | --->
180 |
181 | 61. Rahul Dewan: ii feel, the way to prevent terrorism is to 'raise the level of discourse' on God and Spirituality, and make it spread across all religious faiths".
182 | --->
183 |
184 | 62. And equally like the 19th century, our governing powers and ruling elites don't quite have any answer of how to undo what can't really be undone.
185 | --->
186 |
187 | 63. Commenting on Geof Mosher alongg with Tj Green) Iwould start b ystating the fact that evolutionn is war" is quite an understatment.
188 | --->
189 |
190 | 64. I will address them personally if you wish me to do so.
191 | --->
192 |
193 | 65. That's the wonderful right of freedom.
194 | --->
195 |
196 | 66. Very nice.
197 | --->
198 |
199 | 67. But could the opposite be true: that the gregariousness that is the driving force of human beings, is being virtualized by make-believe substitutes to actual social interaction?
200 | --->
201 |
202 | 68. Guys use Vlc player, the mp4 480p version works perfectly on it, awesome quality too.
203 | --->
204 |
205 | 69. Would it be impolite to ask where you stand in this argument?
206 | --->
207 |
208 | 70. Just wow.
209 | --->
210 |
211 | 71. However, had large numbers of Dutch with their hunting and military style rifles been able to start a sniping campaign against the Germans, Wi may have been shortened.
212 | --->
213 |
214 | 72. Amazing - Good Work - He is Worth to not only to India but also to the Whole world His promise for Open Source is also unbelivable a lots of Thanks to him and sponsers
215 | --->
216 |
217 | 73. It isn't good or bad; it's inevitable.
218 | --->
219 |
220 | 74. A couple of years ago the idea was about privacy that if you are connected to the Internet then your privacy is eliminated and you don't have any right to claim it back because you're a part of the network and you're obviously exchanging info that crosses a lot of paths to reach you, but recently, telecom providers are the ones dominating and watching, closely, what everyone is doing.
221 | --->
222 |
223 | 75. Wait for gasoline to hit $5 a gallon and it will.
224 | --->
225 |
226 | 76. Ib is a rigorous international education program for high school students.
227 | --->
228 |
229 | 77. As someone who has spent several years teaching, I can say from experience that children need discipline, a rigid structure and for the most part, they need to be told what to do.
230 | --->
231 |
232 | 78. Horrible architect, decent marketer perhaps?
233 | --->
234 |
235 | 79. They aren't reproducible on command, and they do not openly discussed the incidences because the subject is taboo.
236 | --->
237 |
238 | 80. Examples: To say that conservatives are much more into ideological purity is absurd.
239 | --->
240 |
241 | 81. David is talking about a job interview on steroids, and like all job interviews there is an understanding that things will be shown in the best light possible (to say the least).
242 | --->
243 |
244 | 82. In order to ensure we aren't fluctuations, we need to make an observation and analyze it.
245 | --->
246 |
247 | 83. They don’t have room or numbers and we are losing them.
248 | --->
249 |
250 | 84. The point I'm trying to make is this: a great craftsman is Not by default a great artist.
251 | --->
252 |
253 | 85. They are invited to raise their right hand and in unison say globall Truce Starts Now!
254 | --->
255 |
256 | 86. I would say all the fundamentalisms floating around in peoples minds making war on each other are one of the great problems of our time.
257 | --->
258 |
259 | 87. I will probably be one of the first to be swept away!
260 | --->
261 |
262 | 88. Please don't even Think of taking it down!
263 | --->
264 |
265 | 89. You see a lot of people, youngsters especially, who complain that they are depressed and alone.
266 | --->
267 |
268 | 90. I'm dubious about the payback periods you state.
269 | --->
270 |
271 | 91. Did his disgraceful comment on Israel fall on deaf ears?
272 | --->
273 |
274 | 92. To that end, how would empathy fix any of these problems?
275 | --->
276 |
277 | 93. follow his lead by taking a hard look at your own life and either get help or work on your own to become more 'gay'.
278 | --->
279 |
280 | 94. Sorry, he's a pet peeve.
281 | --->
282 |
283 | 95. human have no right and ability to make it!
284 | --->
285 |
286 | 96. Felix, why should an artist or a political activist not be commercially astute?
287 | --->
288 |
289 | 97. And now we have screwed up our heads were screwing about with monkeys heads as well):
290 | --->
291 |
292 | 98. It searches for and indexes common data in a vast resource and, not surprisingly, ends up finding lots of that data.
293 | --->
294 |
295 | 99. People need to try to be less offended by differing viewpoints.
296 | --->
297 |
298 | 100. their use of memetics suggests an intentionalityy" in memes.
299 | --->
300 |
301 | 101. Noah Feldman constantly refers to the supposed democratic beliefs of a vast majority of Muslims.
302 | --->
303 |
304 | 102. I felt it helped me deal with issues in a good way and was the right attitude to have.
305 | --->
306 |
307 | 103. This feed specific microbes that in tern secrete the nutrient that the plant needs.
308 | --->
309 |
310 | 104. I know that's a bit simplistic, but still.
311 | --->
312 |
313 | 105. another one thing is staffs, they work only for salary they wouldn't give clear leacture.
314 | --->
315 |
316 | 106. Ingenious incorporation of the natural binding aspect of the mycelium from the mushroom plant to create a products)s that could partially compete with some single uses of Styrofoam.
317 | --->
318 |
319 | 107. As for China, I mostly agree with Mr. Tabarok, but I'm not sure if the post-mao development would've been possible without Mao's prior contribution to health seee Hans Rosling's Ted talks).
320 | --->
321 |
322 | 108. his talk is great.
323 | --->
324 |
325 | 109. Now, to briefly address your list of supposed contradictions and fallacies.
326 | --->
327 |
328 | 110. Lol - hilarious.
329 | --->
330 |
331 | 111. But the point is missed what Dan Gilbert is trying to tell you how we proceed looking at some time of value because its illusion, mostly thats what we are buying "that Illusion!
332 | --->
333 |
334 | 112. Stephen Colbert, Jon Stewart and Bill Maher do make things easier to discuss with humor but they talk about American issues and Americans watch them.
335 | --->
336 |
337 | 113. xd It's been awhile since I smiled so much just from a video ^ I want more of this!
338 | --->
339 |
340 | 114. In my view , school is really a good place to sort out the smart ones who survive the education system and create stupid people (to work as labour) .
341 | --->
342 |
343 | 115. Having been a professional background vocalist in Hollywood for several decades, I att age 24) rejoined a pop band, which had drastically changed its repetoire during my absence, having added several vocally demanding rock tunes, which I would be required to sing to regain my old position.
344 | --->
345 |
346 | 116. Thank goodness and for the interactive transcript feature and Lee Smolin.
347 | --->
348 |
349 | 117. I initiated and developed this project some years ago, which makes it possible for disabled people to take a look beyond their horizons and experience the ultimate feeling of freedom and independence: htp:/w.youtube.com/watch?v=ljbmau7jc
350 | --->
351 |
352 | 118. Pinker uses warped statistics ass pointed out in the blog you mention).
353 | --->
354 |
355 | 119. it's also our fault too by allowing men to do so.
356 | --->
357 |
358 | 120. many are institutionalised profit based or religious hagovers from another world ago.
359 | --->
360 |
361 | 121. This is a political manipulation attempt to educate the children.
362 | --->
363 |
364 | 122. It's a glorified 70-year-old hand cannon and it is not modern, civil, humane or precise.
365 | --->
366 |
367 | 123. Peace!
368 | --->
369 |
370 | 124. Of course this may not guarantee.
371 | --->
372 |
373 | 125. This was some kind of parody, right?
374 | --->
375 |
376 | 126. I do these things for a distraction and generally when I am bored.
377 | --->
378 |
379 | 127. right or wrong I think this is an inevitable advancement you have to be fine with it.
380 | --->
381 |
382 | 128. Would really love to go to a postsecret live event.
383 | --->
384 |
385 | 129. " the words from a man, whom I think, seriously meant that.
386 | --->
387 |
388 | 130. I can see that I wasn't as clear in my writing as I thought I was.
389 | --->
390 |
391 | 131. tragic at the least and still playing out every day.
392 | --->
393 |
394 | 132. I wonder how well this was received and how many of his countrymen saw this.
395 | --->
396 |
397 | 133. I am no scientist but it would seem possible to manipulate such large amounts of data on societal reactions and interactions in a way that would produce highly reliable prognostications relative to very specific stimulus.
398 | --->
399 |
400 | 134. While one size will not fit all, this looks to be a wonderful piece in the puzzle.
401 | --->
402 |
403 | 135. any body please help
404 | --->
405 |
406 | 136. This allows our brains to wire for efficiency.
407 | --->
408 |
409 | 137. When I built the house, I built it according to mostt of) the building codes in Canada.
410 | --->
411 |
412 | 138. 2)2 He created them male and female and blessed them and named them [both] Adam [man] at the time they were created.
413 | --->
414 |
415 | 139. I definitely think, that there is no movie, that can be so epic and so truly touching that these real life - stuffs happening here in Ted.
416 | --->
417 |
418 | 140. All reasearch is of course fantastic, and I take your points about satelites seriously, although suspect that most of us could do without sky tv, gps and various military derivatives if we're brutally honest.
419 | --->
420 |
421 |
--------------------------------------------------------------------------------
/TED_comment_annotations/5/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 2043
5 | svn://deepcore.gr/dc/ted/v1/hlt_project/utils/full_corpus/subjects/testing/5
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2013-01-18T18:08:40.681354Z
11 | 1662
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | comment_test.txt
30 | file
31 |
32 |
33 |
34 |
35 | 2013-01-18T19:54:08.000000Z
36 | 9dac5c2b37dda0ef1690c72bc194a66a
37 | 2013-01-18T18:08:40.681354Z
38 | 1662
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 35826
62 |
63 | sentence_test.txt
64 | file
65 |
66 |
67 |
68 |
69 | 2013-01-18T19:54:08.000000Z
70 | 0e4a83b879db676dc8822e18e196b941
71 | 2013-01-18T18:08:40.681354Z
72 | 1662
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 15350
96 |
97 | complete
98 | dir
99 |
100 | nesli
101 | file
102 |
103 |
104 |
105 |
106 | 2013-01-18T19:54:08.000000Z
107 | d41d8cd98f00b204e9800998ecf8427e
108 | 2013-01-18T18:08:40.681354Z
109 | 1662
110 | snupdc
111 |
112 |
113 |
114 |
115 |
116 |
117 |
118 |
119 |
120 |
121 |
122 |
123 |
124 |
125 |
126 |
127 |
128 |
129 |
130 |
131 |
132 | 0
133 |
134 |
--------------------------------------------------------------------------------
/TED_comment_annotations/5/.svn/text-base/nesli.svn-base:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/TED_comment_annotations/5/.svn/text-base/nesli.svn-base
--------------------------------------------------------------------------------
/TED_comment_annotations/5/complete/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 2043
5 | svn://deepcore.gr/dc/ted/v1/hlt_project/utils/full_corpus/subjects/testing/5/complete
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2013-01-18T18:08:40.681354Z
11 | 1662
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | comment_test.txt
30 | file
31 |
32 |
33 |
34 |
35 | 2013-01-18T19:54:08.000000Z
36 | 10df67bb1b966d3e5a86a45def9aa9dd
37 | 2013-01-18T18:08:40.681354Z
38 | 1662
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 35870
62 |
63 | sentence_test.txt
64 | file
65 |
66 |
67 |
68 |
69 | 2013-01-18T19:54:08.000000Z
70 | 960b32687763c69f5dcf2b9b8e2950e4
71 | 2013-01-18T18:08:40.681354Z
72 | 1662
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 15490
96 |
97 |
--------------------------------------------------------------------------------
/TED_comment_annotations/6/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 2043
5 | svn://deepcore.gr/dc/ted/v1/hlt_project/utils/full_corpus/subjects/testing/6
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2013-01-18T18:08:40.681354Z
11 | 1662
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | comment_test.txt
30 | file
31 |
32 |
33 |
34 |
35 | 2013-01-18T19:54:08.000000Z
36 | fe2a2f87b74c9659be22ce8fc7cfcd79
37 | 2013-01-18T18:08:40.681354Z
38 | 1662
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 34289
62 |
63 | sentence_test.txt
64 | file
65 |
66 |
67 |
68 |
69 | 2013-01-18T19:54:08.000000Z
70 | 8534c630c098c3b86035cb588f0ce6d1
71 | 2013-01-18T18:08:40.681354Z
72 | 1662
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 15056
96 |
97 | complete
98 | dir
99 |
100 | subheim
101 | file
102 |
103 |
104 |
105 |
106 | 2013-01-18T19:54:08.000000Z
107 | d41d8cd98f00b204e9800998ecf8427e
108 | 2013-01-18T18:08:40.681354Z
109 | 1662
110 | snupdc
111 |
112 |
113 |
114 |
115 |
116 |
117 |
118 |
119 |
120 |
121 |
122 |
123 |
124 |
125 |
126 |
127 |
128 |
129 |
130 |
131 |
132 | 0
133 |
134 |
--------------------------------------------------------------------------------
/TED_comment_annotations/6/.svn/text-base/subheim.svn-base:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/TED_comment_annotations/6/.svn/text-base/subheim.svn-base
--------------------------------------------------------------------------------
/TED_comment_annotations/6/complete/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 2043
5 | svn://deepcore.gr/dc/ted/v1/hlt_project/utils/full_corpus/subjects/testing/6/complete
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2013-01-18T18:08:40.681354Z
11 | 1662
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | comment_test.txt
30 | file
31 |
32 |
33 |
34 |
35 | 2013-01-18T19:54:08.000000Z
36 | 656bbac049c18c99a919ac37ab8955b7
37 | 2013-01-18T18:08:40.681354Z
38 | 1662
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 34592
62 |
63 | sentence_test.txt
64 | file
65 |
66 |
67 |
68 |
69 | 2013-01-18T19:54:08.000000Z
70 | e4b8259a46324d95d28285e9a9b0e839
71 | 2013-01-18T18:08:40.681354Z
72 | 1662
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 15196
96 |
97 |
--------------------------------------------------------------------------------
/bootstrapping.py:
--------------------------------------------------------------------------------
1 | ####################################################################
2 | # Licence: Creative Commons (see COPYRIGHT) #
3 | # Authors: Nikolaos Pappas, Georgios Katsimpras #
4 | # {nik0spapp, gkatsimpras}@gmail.com #
5 | # Supervisor: Efstathios stamatatos #
6 | # stamatatos@aegean.gr #
7 | # University of the Aegean #
8 | # Department of Information and Communication Systems Engineering #
9 | # Information Management Track (MSc) #
10 | # Karlovasi, Samos #
11 | # Greece #
12 | ####################################################################
13 |
14 | import re
15 | import sys
16 | import pickle
17 | from terminal_colors import Tcolors
18 | from pb_classifiers import PbSubj
19 |
20 | class Bootstrapping:
21 | """
22 | Bootstrapping: Class performing the bootstrapping process for
23 | subjectivity and objectivity classification of sentences. The
24 | method learns linguistically rich extraction patterns for subjective
25 | (opinionated) expressions from unannotated data. The learned
26 | patterns are used to identify more subjective sentences that simple
27 | high precision classifiers can't recall.
28 |
29 | Related paper:
30 | E. Riloff and J. Wiebe. Learning extraction patterns for subjective
31 | expressions. In Proceedings of the 2003 conference on Empirical methods
32 | in natural language processing, EMNLP '03, pages 105--112, 2003. ACL.
33 |
34 | Learned patterns structure
35 | e.g. {" was killed" : {'type' : 'subj',
36 | 'display' : 'was killed',
37 | 'subj_freq' : 10,
38 | 'freq' : 20,
39 | 'prob' : 0.5}}
40 |
41 | """
42 |
43 | def __init__(self, hp_obj, hp_subj, tagger, debug=False):
44 | # Syntactic forms for pattern extraction
45 | self.syntactic_forms = {"subj" : [["BE","VBN*|VBD*"],
46 | ["HAVE","BE","VB*"],
47 | ["VB*"],
48 | ["VB*","*","NN*|NP*|NC*"],
49 | ["VB*","TO","VB*"],
50 | ["HAVE","TO","BE"],
51 | ["HAVE","NN*"]],
52 | "dobj" : [["VB*"],
53 | ["TO","VB*"],
54 | ["VB*","TO","VB*"]],
55 | "np" : [["NN","IN"],
56 | ["VB*","NN","IN"],
57 | ["BE","VBN","IN"],
58 | ["TO","VB","TO"]]
59 | }
60 | self.filename = "stored/learned_patterns"
61 | try:
62 | self.learned_patterns = pickle.load(open(self.filename))
63 | print Tcolors.ADD + Tcolors.OKBLUE + " Loaded existing pattern knowledge!" + Tcolors.ENDC
64 | except:
65 | print Tcolors.ACT + Tcolors.RED + " Existing pattern knowledge not found." + Tcolors.ENDC
66 | self.learned_patterns = {}
67 |
68 | # Part Of Speech Sequential Tagger (Unigram->Bigram->Trigram)
69 | self.tagger = tagger
70 | # Sentence to be classified
71 | self.subjective = False
72 | self.objective = False
73 | # High precision objective classifier
74 | self.hp_obj = hp_obj
75 | # High precision subjective classifier
76 | self.hp_subj = hp_subj
77 | # Pattern-Based Subjective Classifier
78 | self.pb_subj = PbSubj(self.tagger, debug=debug)
79 | # Learned patterns
80 | self.patterns = {}
81 | self.debug = debug
82 |
83 | def classify(self, sentence, previous="", next=""):
84 | """
85 | Subjectivity classification using boostrapping method.
86 | """
87 | # STEP 1: Classify sentence with HP Subjective classifier
88 | self.subjective = self.hp_subj.classify(sentence)
89 | # STEP 1: Get help from learned patterns
90 | if not self.subjective:
91 | if self.debug: print Tcolors.ACT + " Training pattern based classifier...\n"
92 | self.pb_subj.train(self.learned_patterns)
93 | found, self.subjective, obj = self.pb_subj.classify(sentence)
94 |
95 | if not self.subjective and not self.objective:
96 | # STEP 2: Classify sentence with HP Objective classifier
97 | self.objective = self.hp_obj.classify(sentence, previous, next)
98 |
99 | if self.subjective or self.objective:
100 | # STEP 3: Learn
101 | self.learn_patterns_from(sentence)
102 | else:
103 | # STEP 4: Classify based on learned patterns
104 | found, self.subjective, self.objective = self.pb_subj.classify(sentence)
105 | # Uncomment the two following to bootstrap further the subjective
106 | # sentences detected from the pattern-based classifier.
107 | # if self.subjective:
108 | # self.learn_patterns_from(sentence)
109 | if self.subjective:
110 | return 'subjective'
111 | elif self.objective:
112 | return 'objective'
113 | else:
114 | return None
115 |
116 | def learn_patterns_from(self, sentence):
117 | """
118 | Learns extraction patterns associated with subjectivity
119 | from a given sentence.
120 | """
121 | tagged_sentence = self.tagger.tag(sentence)
122 | tags = []
123 | words = []
124 | if self.debug:
125 | print Tcolors.ACT + " Performing part of speech (POS) tagging..." + Tcolors.WARNING
126 | print tagged_sentence
127 | print Tcolors.ENDC
128 | for (w,tag) in tagged_sentence:
129 | if tag is None:
130 | tag = ""
131 | tags.append(tag)
132 | words.append(w)
133 |
134 | self.trigger_patterns(tags, words)
135 |
136 | def match_until_next_nn(self, i, tags, words, form, key):
137 | """
138 | The hard job for triggering the syntactic forms :-)
139 | """
140 | LIMITER = 4
141 | BE = ['was','were','be','being','am','been','are','is']
142 | HAVE = ['has','have','had']
143 | matched = 0
144 | prev_matched = 0
145 | positions_matched = []
146 | learned_pattern = []
147 | star = False
148 |
149 | for j,ctag in enumerate(form):
150 | next = i + j + 1
151 | inner = 0
152 | found = False
153 | while(not found and next < len(tags)):
154 | next += inner
155 | if next < len(words) and ctag == "VB*" and words[next] in HAVE:
156 | next += 1
157 | if next < len(words) and ctag == "VB*" and words[next] in BE:
158 | next += 1
159 | elif next < len(words) and ctag == "VB*" and words[next] in BE:
160 | next += 1
161 | if ctag == "*":
162 | star = True
163 | elif ctag.find("*") > -1:
164 | ortags = ctag.split("|")
165 | for ortag in ortags:
166 | if next < len(tags) and tags[next].find(ortag.replace("*","")) > -1\
167 | and next not in positions_matched:
168 | if star and inner < 2:
169 | matched += 1
170 | matched += 1
171 | positions_matched.append(next)
172 | found = True
173 | elif ctag == "BE":
174 | if next < len(tags) and (tags[next].find("VB") > -1 or tags[next].find("BE") > -1) \
175 | and words[next] in BE and next not in positions_matched:
176 | matched += 1
177 | positions_matched.append(next)
178 | found = True
179 | elif ctag == "HAVE":
180 | if next < len(tags) and (tags[next].find("VB") > -1 or tags[next].find("HV") > -1)\
181 | and words[next] in HAVE and next not in positions_matched:
182 | matched += 1
183 | positions_matched.append(next)
184 | found = True
185 | elif next < len(tags) and tags[next].find(ctag) > -1\
186 | and next not in positions_matched:
187 | matched += 1
188 | positions_matched.append(next)
189 | found = True
190 | else:
191 | found = True
192 | inner += 1
193 |
194 | if key == "subj":
195 | learned_pattern = [""]
196 | for pos in positions_matched:
197 | learned_pattern.append(words[pos])
198 | if key != "subj":
199 | learned_pattern.append("<" + key +">")
200 |
201 | learned_pattern = " ".join(learned_pattern)
202 |
203 | if matched == len(form):
204 | if self.debug:
205 | print Tcolors.ACT + Tcolors.RED + " Form triggered: ", form, Tcolors.ENDC
206 | print "Pattern learned:", learned_pattern
207 | return True, learned_pattern
208 | else:
209 | return False, None
210 |
211 | def proccess_learned_pattern(self, pattern):
212 | """
213 | Add pattern to learned patterns if it doesn't exist else
214 | update its probability.
215 | """
216 | if pattern.find("subj") > -1:
217 | key = "subj"
218 | elif pattern.find("dobj") > -1:
219 | key = "dobj"
220 | else:
221 | key = "np"
222 | cur_subj_freq = 0
223 | if self.subjective:
224 | cur_subj_freq = 1
225 | pkey = pattern
226 | pkey = re.sub(r" | | ","",pkey)
227 | if self.learned_patterns.has_key(pattern):
228 | subj_freq = self.learned_patterns[pattern]['subj_freq'] + cur_subj_freq
229 | freq = self.learned_patterns[pattern]['freq'] + 1
230 | prob = (float)(subj_freq)/(float)(freq)
231 | self.learned_patterns[pattern]['prob'] = prob
232 | self.learned_patterns[pattern]['subj_freq'] = subj_freq
233 | self.learned_patterns[pattern]['freq'] = freq
234 | if self.debug: print Tcolors.ADD + Tcolors.HEADER + " Updating pattern:", pattern, Tcolors.ENDC
235 | else:
236 | subj_freq = 0
237 | freq = 1
238 | subj_freq += cur_subj_freq
239 | prob = (float)(subj_freq)/(float)(freq)
240 | self.learned_patterns[pattern] = {'type': key,
241 | 'display': pkey,
242 | 'freq' : freq,
243 | 'subj_freq' : subj_freq,
244 | 'prob' : prob}
245 | if self.debug: print Tcolors.ADD + Tcolors.CYAN + " Learning pattern:", pattern, Tcolors.ENDC
246 |
247 | def store_knowledge(self):
248 | """
249 | Stored learned patterns for future usage.
250 | """
251 | output = open(self.filename, 'wb')
252 | pickle.dump(self.learned_patterns, output)
253 |
254 |
255 | def trigger_patterns(self, tags, words):
256 | """
257 | Method that triggers syntactic forms and returns the learned
258 | patterns from the triggering.
259 | """
260 | patterns = []
261 | if self.debug: print Tcolors.ACT + " Triggering subjective syntactic forms..."
262 | for key in self.syntactic_forms.keys():
263 | syntactic_forms = self.syntactic_forms[key]
264 | if self.debug: print Tcolors.PROC + Tcolors.GRAY + " Checking form group " + key + "..." + Tcolors.ENDC
265 |
266 | for form in syntactic_forms:
267 | for i,tag in enumerate(tags):
268 | if tag.find("NN") > -1 or tag.find("NP") > -1 \
269 | or tag.find("PR") > -1:
270 | triggered, pattern = self.match_until_next_nn(i, tags, words, form, key)
271 | if pattern is not None and pattern not in patterns:
272 | if self.debug: print Tcolors.ACT + Tcolors.RED + " Form triggered: ", form, Tcolors.ENDC
273 | patterns.append(pattern)
274 | for pattern in patterns:
275 | self.proccess_learned_pattern(pattern)
276 | if self.debug:
277 | print Tcolors.OKBLUE
278 | print self.learned_patterns
279 | print Tcolors.ENDC
280 | self.store_knowledge()
281 |
282 | def train(self, data):
283 | """
284 | Method to train the pattern-based classifier
285 | """
286 | for sentence in data:
287 | self.classify(sentence)
288 |
289 | def clear_learned_data(self):
290 | self.learned_patterns = {}
291 |
292 |
293 | if __name__ == "__main__":
294 | from hp_classifiers import HpObj, HpSubj
295 | from pos import SequentialTagger
296 | hp_obj = HpObj()
297 | hp_subj = HpSubj()
298 | tagger = SequentialTagger()
299 | bootstrapping = Bootstrapping(hp_obj, hp_subj, tagger)
300 | if self.debug:
301 | print bootstrapping.classify(sys.argv[1])
302 |
303 |
304 |
--------------------------------------------------------------------------------
/datasets/.svn/entries:
--------------------------------------------------------------------------------
1 | 10
2 |
3 | dir
4 | 583
5 | svn://deepcore.gr/dc/projects/icrawler/icrawler/mining/learning/datasets
6 | svn://deepcore.gr/dc
7 |
8 |
9 |
10 | 2011-02-01T03:02:12.956261Z
11 | 576
12 | snupdc
13 |
14 |
15 |
16 |
17 |
18 |
19 |
20 |
21 |
22 |
23 |
24 |
25 |
26 |
27 | 5120806d-c870-0410-adaa-e3d35f40fc17
28 |
29 | segmentation_corpus.data
30 | file
31 |
32 |
33 |
34 |
35 | 2011-02-01T03:10:39.216122Z
36 | 2f7d9a17cdff7f6e23493e364c7769df
37 | 2011-02-01T03:02:12.956261Z
38 | 576
39 | snupdc
40 |
41 |
42 |
43 |
44 |
45 |
46 |
47 |
48 |
49 |
50 |
51 |
52 |
53 |
54 |
55 |
56 |
57 |
58 |
59 |
60 |
61 | 12681
62 |
63 | subjclueslen1-HLTEMNLP05.tff
64 | file
65 |
66 |
67 |
68 |
69 | 2011-01-30T01:50:27.916001Z
70 | 407e41c19d80bf84b264e6443fcff16c
71 | 2011-01-30T03:58:08.885718Z
72 | 568
73 | snupdc
74 |
75 |
76 |
77 |
78 |
79 |
80 |
81 |
82 |
83 |
84 |
85 |
86 |
87 |
88 |
89 |
90 |
91 |
92 |
93 |
94 |
95 | 662976
96 |
97 | emoticons_patch.py
98 | file
99 |
100 |
101 |
102 |
103 | 2011-02-01T03:10:39.216122Z
104 | 4f311e51ccfc4756e0a84dda712b7396
105 | 2011-02-01T03:02:12.956261Z
106 | 576
107 | snupdc
108 |
109 |
110 |
111 |
112 |
113 |
114 |
115 |
116 |
117 |
118 |
119 |
120 |
121 |
122 |
123 |
124 |
125 |
126 |
127 |
128 |
129 | 2194
130 |
131 | emoticons.data
132 | file
133 |
134 |
135 |
136 |
137 | 2011-01-30T01:51:34.232001Z
138 | d50ca7abe90b55099fa4726ce5f18e7f
139 | 2011-01-30T03:58:08.885718Z
140 | 568
141 | snupdc
142 |
143 |
144 |
145 |
146 |
147 |
148 |
149 |
150 |
151 |
152 |
153 |
154 |
155 |
156 |
157 |
158 |
159 |
160 |
161 |
162 |
163 | 3950
164 |
165 |
--------------------------------------------------------------------------------
/datasets/.svn/text-base/emoticons.data.svn-base:
--------------------------------------------------------------------------------
1 | positive
2 | :-) :) :o) :] :3 :c) :> =] 8) =) :} :^)
3 | :D C: :-D :D 8D xD XD =D =3 <=3 <=8 :-9
4 | :-* :* <3 <333 =^_^= =>.>= =<_<= =>.<=
5 | \,,/ \m/ \m/\>.\m/\o/ \o/ o/\o (^_^) (^-^) (^ ^) (^.^) ???
6 | d^_^b d-_-b (^3^) :-P :P XP :-p :p =p :-Þ :Þ :-b :b
7 | ;-) ;) *) ;] ;D :)~ :-)>.... B) B-) 8) 8-)
8 | negative
9 | e.e O.e O.e <=3 <=8 8===D 8===B --!-- :-( :( :c :< :[ :{
10 | D: D8 D; D= DX v.v :-O :O O_O o_o 8O OwO O-O 0_o O_o O3O o0o ;o_o; o...o 0w0
11 | c.c C.C :-/ :/ :\ =/ =\ :S :| :-X :X :-# :#
12 | :'( ;*( T_T TT_TT T.T Q.Q Q_Q ;_; >:) >;) >:-)
13 | D:< >:( >:C D-:< >:-( :-@ ;( `_´ 3 :& :u
14 | (>_>) (¬_¬) (-_-) (^_^') ^_^_^') ^^" ^^^_.^') ^^_^^; ^&^^.^;& ^^^; ^^^7
15 | q(;^;)p (O_o) (._.) (,_,) (;_;) (T_T) (T~T) (ToT) (T^T) (X_X) x_x
16 | bad_words
17 | ahole
18 | anus
19 | ash0le
20 | ash0les
21 | asholes
22 | ass
23 | Ass Monkey
24 | Assface
25 | assh0le
26 | assh0lez
27 | asshole
28 | assholes
29 | assholz
30 | asswipe
31 | azzhole
32 | bassterds
33 | bastard
34 | bastards
35 | bastardz
36 | basterds
37 | basterdz
38 | Biatch
39 | bitch
40 | bitches
41 | Blow Job
42 | boffing
43 | butthole
44 | buttwipe
45 | c0ck
46 | c0cks
47 | c0k
48 | Carpet Muncher
49 | cawk
50 | cawks
51 | Clit
52 | cnts
53 | cntz
54 | cock
55 | cockhead
56 | cock-head
57 | cocks
58 | CockSucker
59 | cock-sucker
60 | crap
61 | cum
62 | cunt
63 | cunts
64 | cuntz
65 | dick
66 | dild0
67 | dild0s
68 | dildo
69 | dildos
70 | dilld0
71 | dilld0s
72 | dominatricks
73 | dominatrics
74 | dominatrix
75 | dyke
76 | enema
77 | f u c k
78 | f u c k e r
79 | fag
80 | fag1t
81 | faget
82 | fagg1t
83 | faggit
84 | faggot
85 | fagit
86 | fags
87 | fagz
88 | faig
89 | faigs
90 | fart
91 | flipping the bird
92 | fuck
93 | fucker
94 | fuckin
95 | fucking
96 | fucks
97 | Fudge Packer
98 | fuk
99 | Fukah
100 | Fuken
101 | fuker
102 | Fukin
103 | Fukk
104 | Fukkah
105 | Fukken
106 | Fukker
107 | Fukkin
108 | g00k
109 | gay
110 | gayboy
111 | gaygirl
112 | gays
113 | gayz
114 | God-damned
115 | h00r
116 | h0ar
117 | h0re
118 | hells
119 | hoar
120 | hoor
121 | hoore
122 | jackoff
123 | jap
124 | japs
125 | jerk-off
126 | jisim
127 | jiss
128 | jizm
129 | jizz
130 | knob
131 | knobs
132 | knobz
133 | kunt
134 | kunts
135 | kuntz
136 | Lesbian
137 | Lezzian
138 | Lipshits
139 | Lipshitz
140 | masochist
141 | masokist
142 | massterbait
143 | masstrbait
144 | masstrbate
145 | masterbaiter
146 | masterbate
147 | masterbates
148 | Motha Fucker
149 | Motha Fuker
150 | Motha Fukkah
151 | Motha Fukker
152 | Mother Fucker
153 | Mother Fukah
154 | Mother Fuker
155 | Mother Fukkah
156 | Mother Fukker
157 | mother-fucker
158 | Mutha Fucker
159 | Mutha Fukah
160 | Mutha Fuker
161 | Mutha Fukkah
162 | Mutha Fukker
163 | n1gr
164 | nastt
165 | nigger;
166 | nigur;
167 | niiger;
168 | niigr;
169 | orafis
170 | orgasim;
171 | orgasm
172 | orgasum
173 | oriface
174 | orifice
175 | orifiss
176 | packi
177 | packie
178 | packy
179 | paki
180 | pakie
181 | paky
182 | pecker
183 | peeenus
184 | peeenusss
185 | peenus
186 | peinus
187 | pen1s
188 | penas
189 | penis
190 | penis-breath
191 | penus
192 | penuus
193 | Phuc
194 | Phuck
195 | Phuk
196 | Phuker
197 | Phukker
198 | polac
199 | polack
200 | polak
201 | Poonani
202 | pr1c
203 | pr1ck
204 | pr1k
205 | pusse
206 | pussee
207 | pussy
208 | puuke
209 | puuker
210 | queer
211 | queers
212 | queerz
213 | qweers
214 | qweerz
215 | qweir
216 | recktum
217 | rectum
218 | retard
219 | sadist
220 | scank
221 | schlong
222 | screwing
223 | semen
224 | sex
225 | sexy
226 | Sh!t
227 | sh1t
228 | sh1ter
229 | sh1ts
230 | sh1tter
231 | sh1tz
232 | shit
233 | shits
234 | shitter
235 | Shitty
236 | Shity
237 | shitz
238 | Shyt
239 | Shyte
240 | Shytty
241 | Shyty
242 | skanck
243 | skank
244 | skankee
245 | skankey
246 | skanks
247 | Skanky
248 | slut
249 | sluts
250 | Slutty
251 | slutz
252 | son-of-a-bitch
253 | tit
254 | turd
255 | va1jina
256 | vag1na
257 | vagiina
258 | vagina
259 | vaj1na
260 | vajina
261 | vullva
262 | vulva
263 | w0p
264 | wh00r
265 | wh0re
266 | whore
267 | xrated
268 | xxx
269 | b!+ch
270 | bitch
271 | blowjob
272 | clit
273 | arschloch
274 | fuck
275 | shit
276 | ass
277 | asshole
278 | b!tch
279 | b17ch
280 | b1tch
281 | bastard
282 | bi+ch
283 | boiolas
284 | buceta
285 | c0ck
286 | cawk
287 | chink
288 | cipa
289 | clits
290 | cock
291 | cum
292 | cunt
293 | dildo
294 | dirsa
295 | ejakulate
296 | fatass
297 | fcuk
298 | fuk
299 | fux0r
300 | hoer
301 | hore
302 | jism
303 | kawk
304 | l3itch
305 | l3i+ch
306 | lesbian
307 | masturbate
308 | masterbat*
309 | masterbat3
310 | motherfucker
311 | s.o.b.
312 | mofo
313 | nazi
314 | nigga
315 | nigger
316 | nutsack
317 | phuck
318 | pimpis
319 | pusse
320 | pussy
321 | scrotum
322 | sh!t
323 | shemale
324 | shi+
325 | sh!+
326 | slut
327 | smut
328 | teets
329 | tits
330 | boobs
331 | b00bs
332 | teez
333 | testical
334 | testicle
335 | titt
336 | w00se
337 | jackoff
338 | wank
339 | whoar
340 | whore
341 | *damn
342 | *dyke
343 | *fuck*
344 | *shit*
345 | @$$
346 | amcik
347 | andskota
348 | arse*
349 | assrammer
350 | ayir
351 | bi7ch
352 | bitch*
353 | bollock*
354 | breasts
355 | butt-pirate
356 | cabron
357 | cazzo
358 | chraa
359 | chuj
360 | Cock*
361 | cunt*
362 | d4mn
363 | daygo
364 | dego
365 | dick*
366 | dike*
367 | dupa
368 | dziwka
369 | ejackulate
370 | Ekrem*
371 | Ekto
372 | enculer
373 | faen
374 | fag*
375 | fanculo
376 | fanny
377 | feces
378 | feg
379 | Felcher
380 | ficken
381 | fitt*
382 | Flikker
383 | foreskin
384 | Fotze
385 | Fu(*
386 | fuk*
387 | futkretzn
388 | gay
389 | gook
390 | guiena
391 | h0r
392 | h4x0r
393 | hell
394 | helvete
395 | hoer*
396 | honkey
397 | Huevon
398 | hui
399 | injun
400 | jizz
401 | kanker*
402 | kike
403 | klootzak
404 | kraut
405 | knulle
406 | kuk
407 | kuksuger
408 | Kurac
409 | kurwa
410 | kusi*
411 | kyrpa*
412 | lesbo
413 | mamhoon
414 | masturbat*
415 | merd*
416 | mibun
417 | monkleigh
418 | mouliewop
419 | muie
420 | mulkku
421 | muschi
422 | nazis
423 | nepesaurio
424 | nigger*
425 | orospu
426 | paska*
427 | perse
428 | picka
429 | pierdol*
430 | pillu*
431 | pimmel
432 | piss*
433 | pizda
434 | poontsee
435 | poop
436 | porn
437 | p0rn
438 | pr0n
439 | preteen
440 | pula
441 | pule
442 | puta
443 | puto
444 | qahbeh
445 | queef*
446 | rautenberg
447 | schaffer
448 | scheiss*
449 | schlampe
450 | schmuck
451 | screw
452 | sh!t*
453 | sharmuta
454 | sharmute
455 | shipal
456 | shiz
457 | skribz
458 | skurwysyn
459 | sphencter
460 | spic
461 | spierdalaj
462 | splooge
463 | stfu
464 | suka
465 | b00b*
466 | testicle*
467 | titt*
468 | twat
469 | vittu
470 | wank*
471 | wetback*
472 | wichser
473 | wop*
474 | yed
475 | zabourah
476 |
--------------------------------------------------------------------------------
/datasets/.svn/text-base/emoticons_patch.py.svn-base:
--------------------------------------------------------------------------------
1 | import os
2 | import sys
3 | from scrapy.conf import settings
4 | PATH = settings.get("PATH")+"icrawler/icrawler/mining/learning/"
5 |
6 | p_flag = False
7 | n_flag = False
8 | b_flag = False
9 |
10 | positives = {} #{"word1":{"typ":"", "pos1":"", "priorpolarity":}}
11 | negatives = {} #{"word1":{"type":"", "pos1":"", "priorpolarity":}}
12 | emoticons_file = open(PATH + "datasets/emoticons.data","r")
13 | lines = emoticons_file.readlines()
14 |
15 | def patch_emoticons():
16 | global p_flag, n_flag, b_flag
17 |
18 | for line in lines:
19 | if line.startswith("positive"):
20 | "Parsing Positives...."
21 | p_flag = True
22 | n_flag = False
23 |
24 | elif line.startswith("negative"):
25 | "Parsing Negatives...."
26 | n_flag = True
27 | p_flag = False
28 |
29 | elif line.startswith("bad_words"):
30 | n_flag = False
31 | p_flag = False
32 | b_flag = True
33 |
34 |
35 | if p_flag and not line.startswith("positive"):
36 | p_data = line.split(" ")
37 | for data in p_data:
38 | positives[data] = {"type":"strongsub", "emoticon" : True, "pos1":"anypos", "priorpolarity":"positive"}
39 |
40 | if n_flag and not line.startswith("negative"):
41 | n_data = line.split(" ")
42 | for data in n_data:
43 | negatives[data] = {"type":"strongsub", "emoticon" : True, "pos1":"anypos", "priorpolarity":"negative"}
44 |
45 | if b_flag and not line.startswith("bad_words"):
46 | word = line.replace("\n", "")
47 | negatives[word] = {"type":"strongsub", "pos1":"anypos", "priorpolarity":"negative"}
48 |
49 | #print "POS", len(positives.keys())
50 | #print "NEG", len(negatives.keys())
51 |
52 | totals = dict(positives, **negatives)
53 |
54 | #print "TOTALS ", len(totals)
55 | return totals
56 |
57 | def parse_dataset(filename):
58 |
59 | parsed_data = []
60 | dataset = open(os.path.abspath("") +"/"+ filename,"r")
61 | lines = dataset.readlines()
62 |
63 | for line in lines:
64 | parsed_data.append(line.replace("\n", ""))
65 |
66 | #print len(parsed_data)
67 | return parsed_data
68 |
69 | #patch_emoticons()
70 |
71 |
72 |
73 |
74 |
75 |
76 |
77 |
78 |
--------------------------------------------------------------------------------
/datasets/.svn/text-base/segmentation_corpus.data.svn-base:
--------------------------------------------------------------------------------
1 | article
2 | http://abcnews.go.com/International/abc-news-christiane-amanpour-reports-protestors-fight-freedom/story?id=12806642
3 | http://www.nytimes.com/roomfordebate/2011/01/31/mubaraks-role-and-mideast-peace
4 | http://www.mirror.co.uk/news/top-stories/2011/02/01/fernando-torres-and-andy-carroll-the-two-multi-million-pound-strikers-who-are-worlds-apart-115875-22890025/
5 | http://ibnlive.in.com/generalnewsfeed/news/wall-st-gains-on-earnings-economy-egypt-fears-ease/559860.html
6 | http://soccernet.espn.go.com/news/story/_/id/873998/spurs-miss-out-on-adam-by-minutes?cc=5739
7 | http://www.ajc.com/news/nation-world/judge-obamas-health-overhaul-822829.html
8 | http://www.montrealgazette.com/travel/Ottawa+offers+flights+Canadians+urged+leave+Egypt/4193713/story.html
9 | http://www.bbc.co.uk/news/business-12314656
10 | http://www.bbc.co.uk/news/world-africa-12328506
11 | http://sports.espn.go.com/dallas/nfl/news/story?id=6077650
12 | http://blogs.forbes.com/christopherhelman/2011/01/31/will-egypts-revolution-mean-oil-armageddon/
13 | http://www.palestinechronicle.com/view_article_details.php?id=16603
14 | http://www.thestar.com/opinion/editorialopinion/article/931187--egypt-connected-to-revolution
15 | http://www.euractiv.com/en/global-europe/egypt-revolution-hijacks-eu-ministers-agenda-news-501730
16 | http://www.theglobeandmail.com/news/opinions/opinion/lessons-from-a-revolution-irans-that-is/article1889288/
17 | http://www.shalomlife.com/eng/14557/Is_Syria's_Bashar_al-Assad_Next/
18 | http://www.fastcompany.com/1722492/how-social-media-accelerated-the-uprising-in-egypt
19 | http://nation.com.pk/pakistan-news-newspaper-daily-english-online/Politics/31-Jan-2011/Altaf-sounds-Tunisia-Egypt
20 | http://www.afrol.com/articles/37168
21 | http://www.theprovince.com/news/This+start+revolution/4192112/story.html
22 | http://money.cnn.com/2011/01/31/technology/google_executive_missing_egypt/index.htm
23 | http://www.nytimes.com/2011/02/01/world/middleeast/01palestinians.html?_r=1&partner=rss&emc=rss
24 | http://www.ibtimes.com/articles/106733/20110130/egypt-ian-lustick-protests-hosni-mubarak.htm
25 | http://www.usnews.com/opinion/blogs/mary-kate-cary/2011/01/28/egypt-cuts-the-internet-in-the-face-of-revolution
26 | http://threatpost.com/en_us/blogs/facebook-owned-42-egyptian-web-traffic-blackout-013111
27 | http://www.grist.org/article/2011-01-31-the-egyptian-government-has-been-able-to-shut-down-the-internet-
28 | http://www.orlandosentinel.com/news/opinion/os-ed-egypt-unrest-myword-020111-20110131,0,2543131.story
29 | http://globalgrind.com/channel/news/content/1903560/the-revolution-of-the-facebook-generation-by-michael-de-la-rocha/
30 | http://www.csmonitor.com/Commentary/Editorial-Board-Blog/2011/0131/How-history-might-really-be-made-in-Egypt
31 | http://communities.washingtontimes.com/neighborhood/word-national-interest/2011/jan/31/pity-mubaraks-successor/
32 | http://www.cbc.ca/canada/windsor/story/2011/01/31/wdr-egypt-reaction.html
33 | http://www.ennaharonline.com/en/international/5681.html
34 | http://www.freep.com/article/20110131/NEWS06/110131023/U-M-MSU-students-affected-by-Egypt-unrest
35 | http://www.currentintelligence.net/gulfstream/2011/1/31/the-great-arab-revolution-and-the-gulf-states.html
36 | http://omadeon.wordpress.com/2011/01/31/omar-suleiman-chief-torturer/
37 | http://www.expatinvesting.org/egypt-revolution-analysis/
38 | http://newzeal.blogspot.com/2011/01/revolution-in-egypt-out-of-frying-pan.html
39 | http://politeching.wordpress.com/2011/01/25/egypts-revolution-powered-by-the-internet/
40 | http://www.huffingtonpost.com/magda-abufadil/media-key-to-covering-egy_b_815920.html
41 | http://www.workers.org/2011/world/wwp_statement_0203/
42 | http://abcclio.blogspot.com/2011/01/egypt-revolution-at-hand.html
43 | http://kasamaproject.org/2011/01/29/the-egyptian-revolution-a-very-fine-thing/
44 | http://www.egyptianrevolution.com/2011/01/egypts-security-apparatus/
45 | http://www.socialist.net/egypt-revolution-knows-no-frontiers.htm
46 | http://www.islamidavet.com/english/2011/01/28/egypt-revolution-hard-to-predict/
47 | http://www.lewrockwell.com/blog/lewrw/archives/76683.html
48 | http://politicalsex.me/unrest-in-egypt-revolution-sounds-exhausting/
49 | http://www.voxy.co.nz/politics/egyptian-revolution-finally-breaks-out-down-mubarak/1273/79978
50 | http://theglobalawakening.wordpress.com/2011/01/31/egypt-defense-min-threatens-people/
51 | http://flagpole.com/Weekly/WorldView/WorldView-28Jan11
52 | http://www.isaintel.com/2011/01/06/remnants-of-a-greek-past-image-from-the-future/
53 | http://www.anarkismo.net/article/18536
54 | http://daily.swarthmore.edu/announcements/2011/nationalism-anarchism-reform-understanding-political-islam-from-the-inside-out/
55 | http://www.bakersfield.com/opinion/letters/x864760197/They-owe-victims-an-apology
56 | http://www.deccanherald.com/content/127462/creative-boundary.html
57 | http://www.brecorder.com/news/top-stories/1142748:news.html
58 | http://maceiteagain.wordpress.com/2011/01/28/280111-anarchism-an-introduction/
59 | http://blogs.ubc.ca/ross/2011/01/%E2%80%9Canarchism%E2%80%A6is-a-living-force-within-our-life%E2%80%A6%E2%80%9D-anarchism-education-and-new-possibilities/
60 | http://revolutionradio.org/?p=11320
61 | http://www.whoplanswhom.com/blog/2011/socialist-misonceptions-about-market-anarchism/
62 | http://francoistremblay.wordpress.com/2010/12/13/the-case-for-socialist-intellectual-ownership/
63 | http://dancull.wordpress.com/2011/01/28/these-seeds-wont-be-uprooted/
64 | http://antigerman.wordpress.com/2009/11/27/capitalism-a-love-story/
65 | http://www.carsuk.net/lamborghini-lp550-2-new-entry-level-lamborghini-gallardo/
66 | http://www.autonewscast.com/2011/01/26/lamborghini-presents-the-gallardo-lp-560-4-bicolore/
67 | http://www.hot-screensaver.com/2011/01/05/lamborghini-aventador-to-replace-murcielago-in-2011/
68 | http://www.nybooks.com/blogs/nyrblog/2011/jan/05/end-medvedev-revolution/
69 | http://www.ottawacitizen.com/news/Little+will+come+Egypt+middle+class+revolution/4197834/story.html
70 | http://www.telegraph.co.uk/news/wikileaks-files/libya-wikileaks/8294609/INVITATION-FOR-THE-SECRETARY-TO-ATTEND-40TH-ANNIVERSARY-OF-QADHAFIS-REVOLUTION.html
71 | http://economictimes.indiatimes.com/news/news-by-industry/services/advertising/online-revolution-may-not-be-televised-nikesh-arora/articleshow/7400360.cms
72 | http://www.esquire.com/blogs/politics/julian-assange-wikileaks-revolution-5072412
73 | http://www.thewrap.com/movies/column-post/tunisian-producer-making-thriller-about-man-who-inspired-jasmine-revolution-24342
74 | http://www.iranian.com/main/2011/jan/shishaki-time
75 | http://sports.yahoo.com/nfl/news?slug=lc-ingamebetting013111
76 | http://www.theatlantic.com/technology/archive/2011/01/twitter-it-wont-start-a-revolution-but-it-can-feed-one/70530/
77 | http://liveshots.blogs.foxnews.com/2011/01/31/a-secular-revolution-for-now/
78 | http://www.allaccess.com/net-news/archive/story/86644/miranda-lambert-announces-the-revolution-continues
79 | http://www.newsweek.com/2011/01/30/el-shaheed-the-mysterious-anonymous-behind-egypt-s-revolt.html
80 | http://www.washingtonpost.com/wp-dyn/content/article/2011/01/30/AR2011013003175.html
81 | http://english.peopledaily.com.cn/90001/98705/100621/7277916.html
82 | http://www.kuwaittimes.net/read_news.php?newsid=MTE3MTU4NTk3NA==
83 | http://mwcnews.net/focus/analysis/8354-egyptian-revolution.html
84 | http://www.monitor.co.ug/SpecialReports/Elections/-/859108/1098344/-/jnywf8/-/
85 | http://www.abugidainfo.com/?p=17136
86 | http://www.sfbg.com/pixel_vision/2011/01/31/aerial-revolution
87 | http://lagunaniguel.patch.com/articles/sc-journalist-egyptians-take-charge-of-the-country
88 | http://www.thoughtleader.co.za/candiceholdsworth/2011/01/31/just-about-time-for-a-revolution/
89 | http://print.dailymirror.lk/news/front-page-news/34443.html
90 | http://blog.dawn.com/2011/01/31/you-say-you-want-a-revolution/
91 | http://www.mideastyouth.com/2011/01/31/tunisia%E2%80%99s-intifada-an-uprising-or-a-revolution/
92 | http://links.org.au/node/2128
93 | http://www.presstv.ir/detail/162823.html
94 | http://www.jpost.com/Features/InThespotlight/Article.aspx?id=205799
95 | http://www.rodiaki.gr/article.php?id=82981&catid=1&subcatid=5
96 | http://www.athinorama.gr/digital/games/articles/default.aspx?id=10006
97 | http://www.sport-fm.gr/article/454202
98 | http://www.newsbeast.gr/culture/arthro/115907/null/
99 | http://www.theglobeandmail.com/news/politics/ottawa-notebook/will-house-resume-with-a-big-bang/article1888486/
100 | http://www.newser.com/story/110885/japanese-volcano-may-be-headed-for-big-bang.html
101 | http://www.washingtonpost.com/wp-dyn/content/article/2011/01/27/AR2011012707212.html
102 | http://www.daemonstv.com/2011/01/21/the-big-bang-theory-the-love-car-displacement-review/
103 | http://www.iol.co.za/tonight/tv-radio/heights-and-lows-for-arm-of-the-law-in-the-hood-1.1019028
104 | http://www.reuters.com/article/2011/01/14/us-ratings-idUSTRE70D6C020110114
105 | http://tubepotato.net/miley-cyrus-could-take-part-in-%E2%80%9Cthe-big-bang%E2%80%9D-movie/33836
106 | http://www.mcccagora.com/a-e/monroe-big-band-makes-big-bang-at-meyer-theater-1.1921045
107 | http://timesofindia.indiatimes.com/home/science/Missing-link-between-Big-Bang-and-star-creation-found/articleshow/7233583.cms
108 | http://www.greatandhra.com/viewnews.php?id=26660&cat=1&scat=4
109 | http://uk.eonline.com/uberblog/watch_with_kristin/b219735_scoop_big_bang_theoryrsquos_johnny.html
110 | http://www.thetowntalk.com/article/20110121/SPORTS/101210344
111 | http://www.spottedbylocals.com/barcelona/big-bang-bar/
112 | http://www.milforddailynews.com/mysource/boomers/health/x1896054311/Gary-Brown-The-chocolate-diet
113 | http://homedailynews.com/a-car-with-chocolate-layer/4478/
114 | http://www.mysuburbanlife.com/glenellyn/features/x896128704/Vinegar-mixes-with-chocolate-at-The-Olive-Tap-in-Downers-Grove
115 | http://blog.seattlepi.com/recipesyndicate/archives/237642.asp
116 | http://www.eciks.org/english/lajme.php?action=total_news&main_id=1101
117 | http://ny.eater.com/archives/2011/01/boxed_city_bakery_hot_chocolate_jacques_torres_opens.php
118 | http://dailytrojan.com/2011/01/30/chocolate-milk-more-than-a-simple-snack/
119 | http://blogs.desmoinesregister.com/dmr/index.php/2011/01/31/chocolate-prices-to-rise/
120 | http://www.communitypress-online.com/ArticleDisplay.aspx?e=2955313
121 | http://www.csmonitor.com/The-Culture/Food/Stir-It-Up/2011/0127/It-s-National-Chocolate-Cake-Day
122 | http://www.sify.com/news/chocolate-gift-causes-heart-attacks-to-school-boy-news-international-lb2sEiadfcb.html
123 | http://montgomeryvillage.patch.com/articles/a-week-without-pizza
124 | http://www.thegrocer.co.uk/articles.aspx?page=independentarticle&ID=215470
125 | http://blog.seattlepi.com/nowinesnobs/archives/237231.asp
126 | http://www.irontontribune.com/2011/01/29/chocolate-walk-set-for-saturday-in-portsmouth/
127 | http://www.acorn-online.com/joomla15/lewisboroledger/news/localnews/83906-a-chocolate-valentine-to-benefit-league.html
128 | http://www.internetbits.com/girl-scouts-cut-cookie-varieties/57320/
129 | http://www.bemidjipioneer.com/event/article/id/100025477/group/homepage/
130 | http://www.examiner.net/opinions/opinions_columnists/x536400824/Bolton-Do-you-like-chocolate-You-may-be-fueling-violence
131 | http://www.charlotteobserver.com/2011/01/30/2015080/snow-days-sliding-hot-chocolate.html
132 | http://www.coloradoan.com/article/20110131/BUSINESS/101310309/Espresso-People-comes-to-town
133 | http://www.morningstarpublishing.com/articles/2011/01/31/grand_traverse_insider/news/benzie_area/doc4d47176d011fc889886936.txt
134 | http://www.mediaforfreedom.com/readarticle.php?AID=17161
135 | http://www.sigmalive.com/news/scitech/346031
136 | http://en.rian.ru/columnists/20110131/162392378.html
137 | http://www.newvision.co.ug/D/8/16/745324
138 | http://english.ruvr.ru/2011/01/21/40676788.html
139 |
140 | comments
141 | http://www.fool.com/investing/international/2011/01/31/egypt-the-cost-of-revolution.aspx
142 | http://www.huffingtonpost.com/haroon-moghul/4-reasons-why-egypts-revo_b_815848.html
143 | http://www.businessspectator.com.au/bs.nsf/Article/egypt-revolution-cairo-food-prices-oil-pd20110201-DMRQ4?OpenDocument&src=sph
144 | http://pajamasmedia.com/michaelledeen/2011/01/28/revolution-by-whom-for-what/
145 | http://www.cbc.ca/canada/story/2011/01/31/f-egypt-canadian-voices.html
146 | http://bigthink.com/ideas/26699
147 | http://www.pri.org/business/social-entrepreneurs/facebook-helps-foment-revolution-in-egypt-and-tunisia2586.html
148 | http://www.huffingtonpost.com/haroon-moghul/4-reasons-why-egypts-revo_b_815848.html
149 | http://bensix.wordpress.com/2011/01/26/chaotic-thoughts-on-anarchism/
150 | http://libcom.org/history/anarchism-syndicalism-brighton-late-19th-early-20th-centuries
151 | http://gigaom.com/2011/01/14/was-what-happened-in-tunisia-a-twitter-revolution/
152 | http://www.naplesnews.com/blogs/naples-tea-party/2011/jan/31/time_____4___tea____/
153 | http://media.www.gwhatchet.com/media/storage/paper332/news/2011/01/31/News/Witnessing.A.Revolution-3972096.shtml
154 | http://www.politicsdaily.com/2011/01/31/why-conservatives-differ-on-remaking-the-world/
155 | http://www.connectmidmichigan.com/news/story.aspx?id=573825
156 | http://www.care2.com/greenliving/chocolate-may-reduce-risk-of-heart-failure.html
157 | http://www.vcstar.com/news/2011/jan/30/emptying-lenins-tomb/
158 |
--------------------------------------------------------------------------------
/datasets/__init__.py:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/datasets/__init__.py
--------------------------------------------------------------------------------
/datasets/__init__.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/datasets/__init__.pyc
--------------------------------------------------------------------------------
/datasets/emoticons.data:
--------------------------------------------------------------------------------
1 | positive
2 | :-) :) :o) :] :3 :c) :> =] 8) =) :} :^)
3 | :D C: :-D :D 8D xD XD =D =3 <=3 <=8 :-9
4 | :-* :* <3 <333 =^_^= =>.>= =<_<= =>.<=
5 | \,,/ \m/ \m/\>.\m/\o/ \o/ o/\o (^_^) (^-^) (^ ^) (^.^) ???
6 | d^_^b d-_-b (^3^) :-P :P XP :-p :p =p :-Þ :Þ :-b :b
7 | ;-) ;) *) ;] ;D :)~ :-)>.... B) B-) 8) 8-)
8 | negative
9 | e.e O.e O.e <=3 <=8 8===D 8===B --!-- :-( :( :c :< :[ :{
10 | D: D8 D; D= DX v.v :-O :O O_O o_o 8O OwO O-O 0_o O_o O3O o0o ;o_o; o...o 0w0
11 | c.c C.C :-/ :/ :\ =/ =\ :S :| :-X :X :-# :#
12 | :'( ;*( T_T TT_TT T.T Q.Q Q_Q ;_; >:) >;) >:-)
13 | D:< >:( >:C D-:< >:-( :-@ ;( `_´ 3 :& :u
14 | (>_>) (¬_¬) (-_-) (^_^') ^_^_^') ^^" ^^^_.^') ^^_^^; ^&^^.^;& ^^^; ^^^7
15 | q(;^;)p (O_o) (._.) (,_,) (;_;) (T_T) (T~T) (ToT) (T^T) (X_X) x_x
16 | bad_words
17 | ahole
18 | anus
19 | ash0le
20 | ash0les
21 | asholes
22 | ass
23 | Ass Monkey
24 | Assface
25 | assh0le
26 | assh0lez
27 | asshole
28 | assholes
29 | assholz
30 | asswipe
31 | azzhole
32 | bassterds
33 | bastard
34 | bastards
35 | bastardz
36 | basterds
37 | basterdz
38 | Biatch
39 | bitch
40 | bitches
41 | Blow Job
42 | boffing
43 | butthole
44 | buttwipe
45 | c0ck
46 | c0cks
47 | c0k
48 | Carpet Muncher
49 | cawk
50 | cawks
51 | Clit
52 | cnts
53 | cntz
54 | cock
55 | cockhead
56 | cock-head
57 | cocks
58 | CockSucker
59 | cock-sucker
60 | crap
61 | cum
62 | cunt
63 | cunts
64 | cuntz
65 | dick
66 | dild0
67 | dild0s
68 | dildo
69 | dildos
70 | dilld0
71 | dilld0s
72 | dominatricks
73 | dominatrics
74 | dominatrix
75 | dyke
76 | enema
77 | f u c k
78 | f u c k e r
79 | fag
80 | fag1t
81 | faget
82 | fagg1t
83 | faggit
84 | faggot
85 | fagit
86 | fags
87 | fagz
88 | faig
89 | faigs
90 | fart
91 | flipping the bird
92 | fuck
93 | fucker
94 | fuckin
95 | fucking
96 | fucks
97 | Fudge Packer
98 | fuk
99 | Fukah
100 | Fuken
101 | fuker
102 | Fukin
103 | Fukk
104 | Fukkah
105 | Fukken
106 | Fukker
107 | Fukkin
108 | g00k
109 | gay
110 | gayboy
111 | gaygirl
112 | gays
113 | gayz
114 | God-damned
115 | h00r
116 | h0ar
117 | h0re
118 | hells
119 | hoar
120 | hoor
121 | hoore
122 | jackoff
123 | jap
124 | japs
125 | jerk-off
126 | jisim
127 | jiss
128 | jizm
129 | jizz
130 | knob
131 | knobs
132 | knobz
133 | kunt
134 | kunts
135 | kuntz
136 | Lesbian
137 | Lezzian
138 | Lipshits
139 | Lipshitz
140 | masochist
141 | masokist
142 | massterbait
143 | masstrbait
144 | masstrbate
145 | masterbaiter
146 | masterbate
147 | masterbates
148 | Motha Fucker
149 | Motha Fuker
150 | Motha Fukkah
151 | Motha Fukker
152 | Mother Fucker
153 | Mother Fukah
154 | Mother Fuker
155 | Mother Fukkah
156 | Mother Fukker
157 | mother-fucker
158 | Mutha Fucker
159 | Mutha Fukah
160 | Mutha Fuker
161 | Mutha Fukkah
162 | Mutha Fukker
163 | n1gr
164 | nastt
165 | nigger;
166 | nigur;
167 | niiger;
168 | niigr;
169 | orafis
170 | orgasim;
171 | orgasm
172 | orgasum
173 | oriface
174 | orifice
175 | orifiss
176 | packi
177 | packie
178 | packy
179 | paki
180 | pakie
181 | paky
182 | pecker
183 | peeenus
184 | peeenusss
185 | peenus
186 | peinus
187 | pen1s
188 | penas
189 | penis
190 | penis-breath
191 | penus
192 | penuus
193 | Phuc
194 | Phuck
195 | Phuk
196 | Phuker
197 | Phukker
198 | polac
199 | polack
200 | polak
201 | Poonani
202 | pr1c
203 | pr1ck
204 | pr1k
205 | pusse
206 | pussee
207 | pussy
208 | puuke
209 | puuker
210 | queer
211 | queers
212 | queerz
213 | qweers
214 | qweerz
215 | qweir
216 | recktum
217 | rectum
218 | retard
219 | sadist
220 | scank
221 | schlong
222 | screwing
223 | semen
224 | sex
225 | sexy
226 | Sh!t
227 | sh1t
228 | sh1ter
229 | sh1ts
230 | sh1tter
231 | sh1tz
232 | shit
233 | shits
234 | shitter
235 | Shitty
236 | Shity
237 | shitz
238 | Shyt
239 | Shyte
240 | Shytty
241 | Shyty
242 | skanck
243 | skank
244 | skankee
245 | skankey
246 | skanks
247 | Skanky
248 | slut
249 | sluts
250 | Slutty
251 | slutz
252 | son-of-a-bitch
253 | tit
254 | turd
255 | va1jina
256 | vag1na
257 | vagiina
258 | vagina
259 | vaj1na
260 | vajina
261 | vullva
262 | vulva
263 | w0p
264 | wh00r
265 | wh0re
266 | whore
267 | xrated
268 | xxx
269 | b!+ch
270 | bitch
271 | blowjob
272 | clit
273 | arschloch
274 | fuck
275 | shit
276 | ass
277 | asshole
278 | b!tch
279 | b17ch
280 | b1tch
281 | bastard
282 | bi+ch
283 | boiolas
284 | buceta
285 | c0ck
286 | cawk
287 | chink
288 | cipa
289 | clits
290 | cock
291 | cum
292 | cunt
293 | dildo
294 | dirsa
295 | ejakulate
296 | fatass
297 | fcuk
298 | fuk
299 | fux0r
300 | hoer
301 | hore
302 | jism
303 | kawk
304 | l3itch
305 | l3i+ch
306 | lesbian
307 | masturbate
308 | masterbat*
309 | masterbat3
310 | motherfucker
311 | s.o.b.
312 | mofo
313 | nazi
314 | nigga
315 | nigger
316 | nutsack
317 | phuck
318 | pimpis
319 | pusse
320 | pussy
321 | scrotum
322 | sh!t
323 | shemale
324 | shi+
325 | sh!+
326 | slut
327 | smut
328 | teets
329 | tits
330 | boobs
331 | b00bs
332 | teez
333 | testical
334 | testicle
335 | titt
336 | w00se
337 | jackoff
338 | wank
339 | whoar
340 | whore
341 | *damn
342 | *dyke
343 | *fuck*
344 | *shit*
345 | @$$
346 | amcik
347 | andskota
348 | arse*
349 | assrammer
350 | ayir
351 | bi7ch
352 | bitch*
353 | bollock*
354 | breasts
355 | butt-pirate
356 | cabron
357 | cazzo
358 | chraa
359 | chuj
360 | Cock*
361 | cunt*
362 | d4mn
363 | daygo
364 | dego
365 | dick*
366 | dike*
367 | dupa
368 | dziwka
369 | ejackulate
370 | Ekrem*
371 | Ekto
372 | enculer
373 | faen
374 | fag*
375 | fanculo
376 | fanny
377 | feces
378 | feg
379 | Felcher
380 | ficken
381 | fitt*
382 | Flikker
383 | foreskin
384 | Fotze
385 | Fu(*
386 | fuk*
387 | futkretzn
388 | gay
389 | gook
390 | guiena
391 | h0r
392 | h4x0r
393 | hell
394 | helvete
395 | hoer*
396 | honkey
397 | Huevon
398 | hui
399 | injun
400 | jizz
401 | kanker*
402 | kike
403 | klootzak
404 | kraut
405 | knulle
406 | kuk
407 | kuksuger
408 | Kurac
409 | kurwa
410 | kusi*
411 | kyrpa*
412 | lesbo
413 | mamhoon
414 | masturbat*
415 | merd*
416 | mibun
417 | monkleigh
418 | mouliewop
419 | muie
420 | mulkku
421 | muschi
422 | nazis
423 | nepesaurio
424 | nigger*
425 | orospu
426 | paska*
427 | perse
428 | picka
429 | pierdol*
430 | pillu*
431 | pimmel
432 | piss*
433 | pizda
434 | poontsee
435 | poop
436 | porn
437 | p0rn
438 | pr0n
439 | preteen
440 | pula
441 | pule
442 | puta
443 | puto
444 | qahbeh
445 | queef*
446 | rautenberg
447 | schaffer
448 | scheiss*
449 | schlampe
450 | schmuck
451 | screw
452 | sh!t*
453 | sharmuta
454 | sharmute
455 | shipal
456 | shiz
457 | skribz
458 | skurwysyn
459 | sphencter
460 | spic
461 | spierdalaj
462 | splooge
463 | stfu
464 | suka
465 | b00b*
466 | testicle*
467 | titt*
468 | twat
469 | vittu
470 | wank*
471 | wetback*
472 | wichser
473 | wop*
474 | yed
475 | zabourah
476 |
--------------------------------------------------------------------------------
/datasets/emoticons_patch.py:
--------------------------------------------------------------------------------
1 | ####################################################################
2 | # Licence: Creative Commons (see COPYRIGHT) #
3 | # Authors: Nikolaos Pappas, Georgios Katsimpras #
4 | # {nik0spapp, gkatsimpras}@gmail.com #
5 | # Supervisor: Efstathios stamatatos #
6 | # stamatatos@aegean.gr #
7 | # University of the Aegean #
8 | # Department of Information and Communication Systems Engineering #
9 | # Information Management Track (MSc) #
10 | # Karlovasi, Samos #
11 | # Greece #
12 | ####################################################################
13 |
14 | import os
15 | import sys
16 |
17 | p_flag = False
18 | n_flag = False
19 | b_flag = False
20 |
21 | positives = {} #{"word1":{"typ":"", "pos1":"", "priorpolarity":}}
22 | negatives = {} #{"word1":{"type":"", "pos1":"", "priorpolarity":}}
23 | emoticons_file = open("datasets/emoticons.data","r")
24 | lines = emoticons_file.readlines()
25 |
26 | def patch_emoticons():
27 | global p_flag, n_flag, b_flag
28 |
29 | for line in lines:
30 | if line.startswith("positive"):
31 | print "Parsing Positives...."
32 | p_flag = True
33 | n_flag = False
34 |
35 | elif line.startswith("negative"):
36 | print "Parsing Negatives...."
37 | n_flag = True
38 | p_flag = False
39 |
40 | elif line.startswith("bad_words"):
41 | n_flag = False
42 | p_flag = False
43 | b_flag = True
44 |
45 | if p_flag and not line.startswith("positive"):
46 | p_data = line.split(" ")
47 | for data in p_data:
48 | positives[data] = {"type":"strongsub", "emoticon" : True, "pos1":"anypos", "priorpolarity":"positive"}
49 |
50 | if n_flag and not line.startswith("negative"):
51 | n_data = line.split(" ")
52 | for data in n_data:
53 | negatives[data] = {"type":"strongsub", "emoticon" : True, "pos1":"anypos", "priorpolarity":"negative"}
54 |
55 | if b_flag and not line.startswith("bad_words"):
56 | word = line.replace("\n", "")
57 | negatives[word] = {"type":"strongsub", "pos1":"anypos", "priorpolarity":"negative"}
58 | totals = dict(positives, **negatives)
59 | return totals
60 |
61 | def parse_dataset(filename):
62 | parsed_data = []
63 | dataset = open(os.path.abspath("") +"/"+ filename,"r")
64 | lines = dataset.readlines()
65 |
66 | for line in lines:
67 | parsed_data.append(line.replace("\n", ""))
68 | return parsed_data
69 |
70 |
71 |
72 |
73 |
74 |
75 |
76 |
77 |
--------------------------------------------------------------------------------
/datasets/emoticons_patch.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/datasets/emoticons_patch.pyc
--------------------------------------------------------------------------------
/datasets/segmentation_corpus.data:
--------------------------------------------------------------------------------
1 | article
2 | http://abcnews.go.com/International/abc-news-christiane-amanpour-reports-protestors-fight-freedom/story?id=12806642
3 | http://www.nytimes.com/roomfordebate/2011/01/31/mubaraks-role-and-mideast-peace
4 | http://www.mirror.co.uk/news/top-stories/2011/02/01/fernando-torres-and-andy-carroll-the-two-multi-million-pound-strikers-who-are-worlds-apart-115875-22890025/
5 | http://ibnlive.in.com/generalnewsfeed/news/wall-st-gains-on-earnings-economy-egypt-fears-ease/559860.html
6 | http://soccernet.espn.go.com/news/story/_/id/873998/spurs-miss-out-on-adam-by-minutes?cc=5739
7 | http://www.ajc.com/news/nation-world/judge-obamas-health-overhaul-822829.html
8 | http://www.montrealgazette.com/travel/Ottawa+offers+flights+Canadians+urged+leave+Egypt/4193713/story.html
9 | http://www.bbc.co.uk/news/business-12314656
10 | http://www.bbc.co.uk/news/world-africa-12328506
11 | http://sports.espn.go.com/dallas/nfl/news/story?id=6077650
12 | http://blogs.forbes.com/christopherhelman/2011/01/31/will-egypts-revolution-mean-oil-armageddon/
13 | http://www.palestinechronicle.com/view_article_details.php?id=16603
14 | http://www.thestar.com/opinion/editorialopinion/article/931187--egypt-connected-to-revolution
15 | http://www.euractiv.com/en/global-europe/egypt-revolution-hijacks-eu-ministers-agenda-news-501730
16 | http://www.theglobeandmail.com/news/opinions/opinion/lessons-from-a-revolution-irans-that-is/article1889288/
17 | http://www.shalomlife.com/eng/14557/Is_Syria's_Bashar_al-Assad_Next/
18 | http://www.fastcompany.com/1722492/how-social-media-accelerated-the-uprising-in-egypt
19 | http://nation.com.pk/pakistan-news-newspaper-daily-english-online/Politics/31-Jan-2011/Altaf-sounds-Tunisia-Egypt
20 | http://www.afrol.com/articles/37168
21 | http://www.theprovince.com/news/This+start+revolution/4192112/story.html
22 | http://money.cnn.com/2011/01/31/technology/google_executive_missing_egypt/index.htm
23 | http://www.nytimes.com/2011/02/01/world/middleeast/01palestinians.html?_r=1&partner=rss&emc=rss
24 | http://www.ibtimes.com/articles/106733/20110130/egypt-ian-lustick-protests-hosni-mubarak.htm
25 | http://www.usnews.com/opinion/blogs/mary-kate-cary/2011/01/28/egypt-cuts-the-internet-in-the-face-of-revolution
26 | http://threatpost.com/en_us/blogs/facebook-owned-42-egyptian-web-traffic-blackout-013111
27 | http://www.grist.org/article/2011-01-31-the-egyptian-government-has-been-able-to-shut-down-the-internet-
28 | http://www.orlandosentinel.com/news/opinion/os-ed-egypt-unrest-myword-020111-20110131,0,2543131.story
29 | http://globalgrind.com/channel/news/content/1903560/the-revolution-of-the-facebook-generation-by-michael-de-la-rocha/
30 | http://www.csmonitor.com/Commentary/Editorial-Board-Blog/2011/0131/How-history-might-really-be-made-in-Egypt
31 | http://communities.washingtontimes.com/neighborhood/word-national-interest/2011/jan/31/pity-mubaraks-successor/
32 | http://www.cbc.ca/canada/windsor/story/2011/01/31/wdr-egypt-reaction.html
33 | http://www.ennaharonline.com/en/international/5681.html
34 | http://www.freep.com/article/20110131/NEWS06/110131023/U-M-MSU-students-affected-by-Egypt-unrest
35 | http://www.currentintelligence.net/gulfstream/2011/1/31/the-great-arab-revolution-and-the-gulf-states.html
36 | http://omadeon.wordpress.com/2011/01/31/omar-suleiman-chief-torturer/
37 | http://www.expatinvesting.org/egypt-revolution-analysis/
38 | http://newzeal.blogspot.com/2011/01/revolution-in-egypt-out-of-frying-pan.html
39 | http://politeching.wordpress.com/2011/01/25/egypts-revolution-powered-by-the-internet/
40 | http://www.huffingtonpost.com/magda-abufadil/media-key-to-covering-egy_b_815920.html
41 | http://www.workers.org/2011/world/wwp_statement_0203/
42 | http://abcclio.blogspot.com/2011/01/egypt-revolution-at-hand.html
43 | http://kasamaproject.org/2011/01/29/the-egyptian-revolution-a-very-fine-thing/
44 | http://www.egyptianrevolution.com/2011/01/egypts-security-apparatus/
45 | http://www.socialist.net/egypt-revolution-knows-no-frontiers.htm
46 | http://www.islamidavet.com/english/2011/01/28/egypt-revolution-hard-to-predict/
47 | http://www.lewrockwell.com/blog/lewrw/archives/76683.html
48 | http://politicalsex.me/unrest-in-egypt-revolution-sounds-exhausting/
49 | http://www.voxy.co.nz/politics/egyptian-revolution-finally-breaks-out-down-mubarak/1273/79978
50 | http://theglobalawakening.wordpress.com/2011/01/31/egypt-defense-min-threatens-people/
51 | http://flagpole.com/Weekly/WorldView/WorldView-28Jan11
52 | http://www.isaintel.com/2011/01/06/remnants-of-a-greek-past-image-from-the-future/
53 | http://www.anarkismo.net/article/18536
54 | http://daily.swarthmore.edu/announcements/2011/nationalism-anarchism-reform-understanding-political-islam-from-the-inside-out/
55 | http://www.bakersfield.com/opinion/letters/x864760197/They-owe-victims-an-apology
56 | http://www.deccanherald.com/content/127462/creative-boundary.html
57 | http://www.brecorder.com/news/top-stories/1142748:news.html
58 | http://maceiteagain.wordpress.com/2011/01/28/280111-anarchism-an-introduction/
59 | http://blogs.ubc.ca/ross/2011/01/%E2%80%9Canarchism%E2%80%A6is-a-living-force-within-our-life%E2%80%A6%E2%80%9D-anarchism-education-and-new-possibilities/
60 | http://revolutionradio.org/?p=11320
61 | http://www.whoplanswhom.com/blog/2011/socialist-misonceptions-about-market-anarchism/
62 | http://francoistremblay.wordpress.com/2010/12/13/the-case-for-socialist-intellectual-ownership/
63 | http://dancull.wordpress.com/2011/01/28/these-seeds-wont-be-uprooted/
64 | http://antigerman.wordpress.com/2009/11/27/capitalism-a-love-story/
65 | http://www.carsuk.net/lamborghini-lp550-2-new-entry-level-lamborghini-gallardo/
66 | http://www.autonewscast.com/2011/01/26/lamborghini-presents-the-gallardo-lp-560-4-bicolore/
67 | http://www.hot-screensaver.com/2011/01/05/lamborghini-aventador-to-replace-murcielago-in-2011/
68 | http://www.nybooks.com/blogs/nyrblog/2011/jan/05/end-medvedev-revolution/
69 | http://www.ottawacitizen.com/news/Little+will+come+Egypt+middle+class+revolution/4197834/story.html
70 | http://www.telegraph.co.uk/news/wikileaks-files/libya-wikileaks/8294609/INVITATION-FOR-THE-SECRETARY-TO-ATTEND-40TH-ANNIVERSARY-OF-QADHAFIS-REVOLUTION.html
71 | http://economictimes.indiatimes.com/news/news-by-industry/services/advertising/online-revolution-may-not-be-televised-nikesh-arora/articleshow/7400360.cms
72 | http://www.esquire.com/blogs/politics/julian-assange-wikileaks-revolution-5072412
73 | http://www.thewrap.com/movies/column-post/tunisian-producer-making-thriller-about-man-who-inspired-jasmine-revolution-24342
74 | http://www.iranian.com/main/2011/jan/shishaki-time
75 | http://sports.yahoo.com/nfl/news?slug=lc-ingamebetting013111
76 | http://www.theatlantic.com/technology/archive/2011/01/twitter-it-wont-start-a-revolution-but-it-can-feed-one/70530/
77 | http://liveshots.blogs.foxnews.com/2011/01/31/a-secular-revolution-for-now/
78 | http://www.allaccess.com/net-news/archive/story/86644/miranda-lambert-announces-the-revolution-continues
79 | http://www.newsweek.com/2011/01/30/el-shaheed-the-mysterious-anonymous-behind-egypt-s-revolt.html
80 | http://www.washingtonpost.com/wp-dyn/content/article/2011/01/30/AR2011013003175.html
81 | http://english.peopledaily.com.cn/90001/98705/100621/7277916.html
82 | http://www.kuwaittimes.net/read_news.php?newsid=MTE3MTU4NTk3NA==
83 | http://mwcnews.net/focus/analysis/8354-egyptian-revolution.html
84 | http://www.monitor.co.ug/SpecialReports/Elections/-/859108/1098344/-/jnywf8/-/
85 | http://www.abugidainfo.com/?p=17136
86 | http://www.sfbg.com/pixel_vision/2011/01/31/aerial-revolution
87 | http://lagunaniguel.patch.com/articles/sc-journalist-egyptians-take-charge-of-the-country
88 | http://www.thoughtleader.co.za/candiceholdsworth/2011/01/31/just-about-time-for-a-revolution/
89 | http://print.dailymirror.lk/news/front-page-news/34443.html
90 | http://blog.dawn.com/2011/01/31/you-say-you-want-a-revolution/
91 | http://www.mideastyouth.com/2011/01/31/tunisia%E2%80%99s-intifada-an-uprising-or-a-revolution/
92 | http://links.org.au/node/2128
93 | http://www.presstv.ir/detail/162823.html
94 | http://www.jpost.com/Features/InThespotlight/Article.aspx?id=205799
95 | http://www.rodiaki.gr/article.php?id=82981&catid=1&subcatid=5
96 | http://www.athinorama.gr/digital/games/articles/default.aspx?id=10006
97 | http://www.sport-fm.gr/article/454202
98 | http://www.newsbeast.gr/culture/arthro/115907/null/
99 | http://www.theglobeandmail.com/news/politics/ottawa-notebook/will-house-resume-with-a-big-bang/article1888486/
100 | http://www.newser.com/story/110885/japanese-volcano-may-be-headed-for-big-bang.html
101 | http://www.washingtonpost.com/wp-dyn/content/article/2011/01/27/AR2011012707212.html
102 | http://www.daemonstv.com/2011/01/21/the-big-bang-theory-the-love-car-displacement-review/
103 | http://www.iol.co.za/tonight/tv-radio/heights-and-lows-for-arm-of-the-law-in-the-hood-1.1019028
104 | http://www.reuters.com/article/2011/01/14/us-ratings-idUSTRE70D6C020110114
105 | http://tubepotato.net/miley-cyrus-could-take-part-in-%E2%80%9Cthe-big-bang%E2%80%9D-movie/33836
106 | http://www.mcccagora.com/a-e/monroe-big-band-makes-big-bang-at-meyer-theater-1.1921045
107 | http://timesofindia.indiatimes.com/home/science/Missing-link-between-Big-Bang-and-star-creation-found/articleshow/7233583.cms
108 | http://www.greatandhra.com/viewnews.php?id=26660&cat=1&scat=4
109 | http://uk.eonline.com/uberblog/watch_with_kristin/b219735_scoop_big_bang_theoryrsquos_johnny.html
110 | http://www.thetowntalk.com/article/20110121/SPORTS/101210344
111 | http://www.spottedbylocals.com/barcelona/big-bang-bar/
112 | http://www.milforddailynews.com/mysource/boomers/health/x1896054311/Gary-Brown-The-chocolate-diet
113 | http://homedailynews.com/a-car-with-chocolate-layer/4478/
114 | http://www.mysuburbanlife.com/glenellyn/features/x896128704/Vinegar-mixes-with-chocolate-at-The-Olive-Tap-in-Downers-Grove
115 | http://blog.seattlepi.com/recipesyndicate/archives/237642.asp
116 | http://www.eciks.org/english/lajme.php?action=total_news&main_id=1101
117 | http://ny.eater.com/archives/2011/01/boxed_city_bakery_hot_chocolate_jacques_torres_opens.php
118 | http://dailytrojan.com/2011/01/30/chocolate-milk-more-than-a-simple-snack/
119 | http://blogs.desmoinesregister.com/dmr/index.php/2011/01/31/chocolate-prices-to-rise/
120 | http://www.communitypress-online.com/ArticleDisplay.aspx?e=2955313
121 | http://www.csmonitor.com/The-Culture/Food/Stir-It-Up/2011/0127/It-s-National-Chocolate-Cake-Day
122 | http://www.sify.com/news/chocolate-gift-causes-heart-attacks-to-school-boy-news-international-lb2sEiadfcb.html
123 | http://montgomeryvillage.patch.com/articles/a-week-without-pizza
124 | http://www.thegrocer.co.uk/articles.aspx?page=independentarticle&ID=215470
125 | http://blog.seattlepi.com/nowinesnobs/archives/237231.asp
126 | http://www.irontontribune.com/2011/01/29/chocolate-walk-set-for-saturday-in-portsmouth/
127 | http://www.acorn-online.com/joomla15/lewisboroledger/news/localnews/83906-a-chocolate-valentine-to-benefit-league.html
128 | http://www.internetbits.com/girl-scouts-cut-cookie-varieties/57320/
129 | http://www.bemidjipioneer.com/event/article/id/100025477/group/homepage/
130 | http://www.examiner.net/opinions/opinions_columnists/x536400824/Bolton-Do-you-like-chocolate-You-may-be-fueling-violence
131 | http://www.charlotteobserver.com/2011/01/30/2015080/snow-days-sliding-hot-chocolate.html
132 | http://www.coloradoan.com/article/20110131/BUSINESS/101310309/Espresso-People-comes-to-town
133 | http://www.morningstarpublishing.com/articles/2011/01/31/grand_traverse_insider/news/benzie_area/doc4d47176d011fc889886936.txt
134 | http://www.mediaforfreedom.com/readarticle.php?AID=17161
135 | http://www.sigmalive.com/news/scitech/346031
136 | http://en.rian.ru/columnists/20110131/162392378.html
137 | http://www.newvision.co.ug/D/8/16/745324
138 | http://english.ruvr.ru/2011/01/21/40676788.html
139 |
140 | comments
141 | http://www.fool.com/investing/international/2011/01/31/egypt-the-cost-of-revolution.aspx
142 | http://www.huffingtonpost.com/haroon-moghul/4-reasons-why-egypts-revo_b_815848.html
143 | http://www.businessspectator.com.au/bs.nsf/Article/egypt-revolution-cairo-food-prices-oil-pd20110201-DMRQ4?OpenDocument&src=sph
144 | http://pajamasmedia.com/michaelledeen/2011/01/28/revolution-by-whom-for-what/
145 | http://www.cbc.ca/canada/story/2011/01/31/f-egypt-canadian-voices.html
146 | http://bigthink.com/ideas/26699
147 | http://www.pri.org/business/social-entrepreneurs/facebook-helps-foment-revolution-in-egypt-and-tunisia2586.html
148 | http://www.huffingtonpost.com/haroon-moghul/4-reasons-why-egypts-revo_b_815848.html
149 | http://bensix.wordpress.com/2011/01/26/chaotic-thoughts-on-anarchism/
150 | http://libcom.org/history/anarchism-syndicalism-brighton-late-19th-early-20th-centuries
151 | http://gigaom.com/2011/01/14/was-what-happened-in-tunisia-a-twitter-revolution/
152 | http://www.naplesnews.com/blogs/naples-tea-party/2011/jan/31/time_____4___tea____/
153 | http://media.www.gwhatchet.com/media/storage/paper332/news/2011/01/31/News/Witnessing.A.Revolution-3972096.shtml
154 | http://www.politicsdaily.com/2011/01/31/why-conservatives-differ-on-remaking-the-world/
155 | http://www.connectmidmichigan.com/news/story.aspx?id=573825
156 | http://www.care2.com/greenliving/chocolate-may-reduce-risk-of-heart-failure.html
157 | http://www.vcstar.com/news/2011/jan/30/emptying-lenins-tomb/
158 |
--------------------------------------------------------------------------------
/hp_classifiers.py:
--------------------------------------------------------------------------------
1 | ####################################################################
2 | # Licence: Creative Commons (see COPYRIGHT) #
3 | # Authors: Nikolaos Pappas, Georgios Katsimpras #
4 | # {nik0spapp, gkatsimpras}@gmail.com #
5 | # Supervisor: Efstathios stamatatos #
6 | # stamatatos@aegean.gr #
7 | # University of the Aegean #
8 | # Department of Information and Communication Systems Engineering #
9 | # Information Management Track (MSc) #
10 | # Karlovasi, Samos #
11 | # Greece #
12 | ####################################################################
13 |
14 | import sys
15 | import nltk
16 | from lexicon import Lexicon
17 | from stemming.porter2 import stem
18 |
19 | class HpSubj:
20 | """
21 | High precision subjective sentence classifier which uses an annotated
22 | lexicon of words as features. It classifies a sentence as subjective
23 | if it contains two or more of the strong subjective clues.
24 | """
25 |
26 | def __init__(self, debug=False):
27 | self.dictionary = Lexicon().words
28 | self.debug = debug
29 |
30 | def classify(self, sentence):
31 | wdict = self.dictionary
32 | words = nltk.word_tokenize(sentence)
33 | strong_subjective_words_count = 0
34 | subjective = False
35 | for word in words:
36 | word = word.lower()
37 | check = [word, stem(word)]
38 | for w in check:
39 | if wdict.has_key(w) and wdict[w]['type'] == 'strongsubj':
40 | strong_subjective_words_count += 1
41 | if strong_subjective_words_count >= 2:
42 | subjective = True
43 | break
44 | return subjective
45 |
46 |
47 | class HpObj:
48 | """
49 | High precision objective sentence classifier which uses an annotated
50 | lexicon as training data. It classifies a sentence as objective if it
51 | doesn't contain along with its previous and next sentence, not even
52 | one strong subjective clue and at most one weak subjective clue.
53 | """
54 |
55 | def __init__(self, debug=False):
56 |
57 | self.lexicon = Lexicon()
58 | self.dictionary = self.lexicon.words
59 | self.debug = debug
60 |
61 | def classify(self, current, previous="", next=""):
62 | if self.debug:
63 | print
64 | print "current:", current
65 | print "previous:", previous
66 | print "next:", next
67 | print
68 | wdict = self.dictionary
69 | words = nltk.word_tokenize(current)
70 | prev_words = nltk.word_tokenize(previous)
71 | next_words = nltk.word_tokenize(next)
72 | words += prev_words + next_words
73 | strong_subjective_words_count = 0
74 | weak_subjective_words_count = 0
75 | objective = True
76 |
77 | for word in words:
78 | word = word.lower()
79 | check = [word, stem(word)]
80 | for w in check:
81 | if wdict.has_key(w):
82 | if wdict[w]['type'] == 'strongsubj':
83 | strong_subjective_words_count += 1
84 | if strong_subjective_words_count > 0:
85 | objective = False
86 | break
87 | elif wdict[w]['type'] == 'weaksubj':
88 | weak_subjective_words_count += 1
89 | if weak_subjective_words_count > 1:
90 | objective = False
91 | break
92 | return objective
93 |
94 |
95 | if __name__ == '__main__':
96 | hpo = HpObj()
97 | hps = HpSubj()
98 | print "Objective: " + (str)(hpo.classify(sys.argv[1]))
99 | print "Subjective: " + (str)(hps.classify(sys.argv[1]))
100 |
101 |
102 |
103 |
104 |
105 |
106 |
107 |
108 |
109 |
110 |
--------------------------------------------------------------------------------
/images/bootstrap.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/images/bootstrap.png
--------------------------------------------------------------------------------
/lexicon.py:
--------------------------------------------------------------------------------
1 | ####################################################################
2 | # Licence: Creative Commons (see COPYRIGHT) #
3 | # Authors: Nikolaos Pappas, Georgios Katsimpras #
4 | # {nik0spapp, gkatsimpras}@gmail.com #
5 | # Supervisor: Efstathios stamatatos #
6 | # stamatatos@aegean.gr #
7 | # University of the Aegean #
8 | # Department of Information and Communication Systems Engineering #
9 | # Information Management Track (MSc) #
10 | # Karlovasi, Samos #
11 | # Greece #
12 | ####################################################################
13 |
14 | import pickle
15 | import os
16 | import sys
17 | from datasets.emoticons_patch import patch_emoticons
18 |
19 | class Lexicon:
20 | """
21 | Lexicon class loads an annotated dataset of words
22 | that have strong/weak subjectivity and is used to
23 | train the high precision objective and subjective
24 | classifiers.
25 | """
26 |
27 | def __init__(self):
28 | self.filename = "stored/lexicon"
29 | try:
30 | self.words = pickle.load(open(self.filename))
31 | except:
32 | self.words = {}
33 | self.load()
34 | output = open(self.filename, 'wb')
35 | pickle.dump(self.words, output)
36 |
37 | def load(self):
38 | """
39 | The method loads the annotated corpus and extracts the structure
40 | with easy access for the classifiers.
41 | """
42 | dictionary_file = open("datasets/subjclueslen1-HLTEMNLP05.tff","r")
43 | lines = dictionary_file.readlines()
44 |
45 | for line in lines:
46 | attributes = line.split(" ")
47 | for index,attr in enumerate(attributes):
48 | if attr.find('word1') > -1:
49 | word_value = attr.split("=")[1]
50 | attributes[index] = []
51 | break
52 | if self.words.has_key(word_value):
53 | for attr in attributes:
54 | if attr != []:
55 | arr = attr.split("=")
56 | key = arr[0]
57 | if key == "pos1":
58 | pos = self.words[word_value][key]
59 | self.words[word_value][key].append(arr[1])
60 | break
61 | else:
62 | self.words[word_value] = {}
63 | for attr in attributes:
64 | if attr != []:
65 | arr = attr.split("=")
66 | key = arr[0]
67 | if len(arr) > 1:
68 | value = arr[1]
69 | if key == "pos1":
70 | self.words[word_value][key] = [value.replace("\n", "")]
71 | else:
72 | self.words[word_value][key] = value.replace("\n", "")
73 |
74 | self.words = dict(patch_emoticons(), **self.words)
75 |
76 |
77 |
78 |
79 |
--------------------------------------------------------------------------------
/lexicon/negative-words.txt:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/lexicon/negative-words.txt
--------------------------------------------------------------------------------
/pb_classifiers.py:
--------------------------------------------------------------------------------
1 | ####################################################################
2 | # Licence: Creative Commons (see COPYRIGHT) #
3 | # Authors: Nikolaos Pappas, Georgios Katsimpras #
4 | # {nik0spapp, gkatsimpras}@gmail.com #
5 | # Supervisor: Efstathios stamatatos #
6 | # stamatatos@aegean.gr #
7 | # University of the Aegean #
8 | # Department of Information and Communication Systems Engineering #
9 | # Information Management Track (MSc) #
10 | # Karlovasi, Samos #
11 | # Greece #
12 | ####################################################################
13 |
14 | import os
15 | import sys
16 | import nltk
17 | import random
18 | import datetime
19 | from terminal_colors import Tcolors
20 | sys.path.append(os.path.abspath("") + "/../")
21 |
22 | class PbSubj:
23 | """
24 | PbSubj: Pattern-Based subjective sentence classifier which classifies a
25 | sentence as subjective with a probability of the top-matched
26 | pattern among a list of strongly associated with subjectivity
27 | patterns. The selection of these patterns is made using two thresholds
28 | t1 and t2. The patterns whom frequency is greater than t1 and the
29 | subjective frequency greater than t2 are selected.
30 | """
31 |
32 | def __init__(self, tagger, debug=False):
33 | self.tagger = tagger
34 | # Patterns learned from the pattern learner
35 | self.learned_patterns = {}
36 | # Strong subjective patterns
37 | self.ss_patterns = {}
38 | self.sorted_ss_patterns = None
39 | self.t1_threshold = 5 # 3
40 | self.t2_threshold = 1 # 0.9
41 | self.pl_threshold = 25
42 | self.limit = 1
43 | self.debug = debug
44 |
45 | def classify(self, sentence):
46 | """
47 | Classify sentence based on the probabilities of the strongly
48 | associated patterns with subjectivity.
49 | """
50 | found = False
51 | matched_pattern = None
52 | # POS tagging
53 | tagged_sentence = self.tagger.tag(sentence)
54 | words = []
55 | tags = []
56 | for (word, tag) in tagged_sentence:
57 | words.append(word)
58 | if tag is None:
59 | tag = ""
60 | tags.append(tag)
61 |
62 | for (pattern, value) in self.sorted_ss_patterns:
63 | display = value['display']
64 | pattern_type = value['type']
65 | try:
66 | pos_in_sentence = sentence.find(display)
67 | except:
68 | pos_in_sentence = -1
69 |
70 | if pos_in_sentence > -1:
71 | matched_pattern = value
72 | if pattern_type == "subj":
73 | found = self.search_for_subject(display, words, tags)
74 | elif pattern_type in ["dobj", "np"]:
75 | remaining_sentence = sentence[pos_in_sentence:]
76 | found = self.search_for_object(display, words, tags)
77 | if found:
78 | break
79 | if not found:
80 | objective = False
81 | subjective = False
82 | else:
83 | if self.debug: print matched_pattern
84 | random.seed()
85 | if random.uniform(0,1) <= matched_pattern['prob']:
86 | subjective = True
87 | objective = False
88 | if self.debug: print "Probability: " + (str)(matched_pattern['prob'])
89 | else:
90 | objective = True
91 | subjective = False
92 | if self.debug: print "Probability: " + (str)(1 - matched_pattern['prob'])
93 | return found, subjective, objective
94 |
95 |
96 | def find_needle_in_haystack(self, needle, haystack):
97 | """
98 | This method finds the position of the intersection on
99 | the haystack array (if there is one).
100 | """
101 | r = []
102 | L = len(needle)
103 | for i in range(len(haystack)):
104 | if haystack[i:i+L] == needle:
105 | r.append(i)
106 | return r
107 |
108 |
109 | def search_for_object(self, pattern, words, tags):
110 | pattern_words = pattern.split()
111 | position = self.find_needle_in_haystack(pattern_words, words)
112 | if len(position) > 0:
113 | position = position[0] + len(pattern_words)
114 | for i, tag in enumerate(tags[position:]):
115 | if i < self.limit and (tag.find("NN") > -1 or tag.find("NP") > -1\
116 | or tag.find("PR") > -1):
117 | return True
118 | return False
119 |
120 |
121 | def search_for_subject(self, pattern, words, tags):
122 | pattern_words = pattern.split()
123 | position = self.find_needle_in_haystack(pattern_words, words)
124 | if len(position) > 0:
125 | position = position[0] - 1
126 | for i, tag in enumerate(tags[position:]):
127 | if i < self.limit and (tag.find("NN") > -1 or tag.find("NP") > -1\
128 | or tag.find("PR") > -1):
129 | return True
130 | return False
131 |
132 |
133 | def select_strong_subjective_patterns(self):
134 | """
135 | Selection of the strongly associated with subjectivity patterns
136 | using the thresholds t1 and t2.
137 | """
138 | self.ss_patterns = {}
139 | for pattern in self.learned_patterns.keys():
140 | freq = self.learned_patterns[pattern]['freq']
141 | prob = self.learned_patterns[pattern]['prob']
142 | if freq >= self.t1_threshold and prob >= self.t2_threshold:
143 | self.ss_patterns[pattern] = self.learned_patterns[pattern]
144 | # delete some patterns with low frequency and probability for efficiency
145 | elif freq > 5 and freq < ((self.t1_threshold*3) / 4):
146 | del(self.learned_patterns[pattern])
147 |
148 | sorted_ss = sorted(self.ss_patterns.iteritems(),key=lambda x: x[1]['prob'], reverse=True)
149 | self.sorted_ss_patterns = sorted_ss
150 | for (s,v) in sorted_ss:
151 | title = (Tcolors.OKGREEN+s+Tcolors.ENDC+" ").ljust(70,'-')
152 | pbs = (str)(v['freq'])+"/" + Tcolors.CYAN + (str)(v['prob']) + Tcolors.ENDC
153 | if self.debug: print title + "------------> " + pbs
154 | if self.debug: print
155 | if len(sorted_ss) > self.pl_threshold:
156 | self.t1_threshold += 1
157 |
158 | def train(self, learned_patterns):
159 | """
160 | Train classifier with the learned patterns derived from
161 | the pattern learner.
162 | """
163 | self.learned_patterns = learned_patterns
164 | self.select_strong_subjective_patterns()
165 |
166 |
167 |
--------------------------------------------------------------------------------
/polarity.py:
--------------------------------------------------------------------------------
1 | ####################################################################
2 | # Licence: Creative Commons (see COPYRIGHT) #
3 | # Authors: Nikolaos Pappas, Georgios Katsimpras #
4 | # {nik0spapp, gkatsimpras}@gmail.com #
5 | # Supervisor: Efstathios stamatatos #
6 | # stamatatos@aegean.gr #
7 | # University of the Aegean #
8 | # Department of Information and Communication Systems Engineering #
9 | # Information Management Track (MSc) #
10 | # Karlovasi, Samos #
11 | # Greece #
12 | ####################################################################
13 |
14 | from __future__ import division
15 | import nltk
16 | import string
17 | import sys
18 | from terminal_colors import Tcolors
19 | from stemming.porter2 import stem
20 | sys.path.append(sys.path[0] + "/../")
21 |
22 |
23 | class PolarityClassifier:
24 | """
25 | PolarityClassifier: Rule-based polarity classification of sentences
26 | according to the following paper:
27 |
28 | T. Wilson, J. Wiebe, and P. Hoffmann. Recognizing contextual polarity
29 | in phrase-level sentiment analysis. In Proceedings of the conference
30 | on Human Language Technology and Empirical Methods in Natural Language
31 | Processing, HLT '05, pages 347--354, 2005.
32 |
33 | Enhancements: We have incorporated emoticons and slung dictionary
34 | apart from the MPQA lexicon that is used in the paper.
35 | """
36 |
37 | def __init__(self, tagger, lexicon, debug=False):
38 | self.lexicon = lexicon.words
39 | self.sentence = None
40 | self.words = []
41 | self.feature_words = {}
42 | self.polar_expressions = []
43 | self.polar_with_tags = {}
44 | self.polar_with_score = {}
45 | self.strong_polar_expressions = []
46 | self.negation_words = ["not", "no", "but"]
47 | self.tagger = tagger
48 | self.words_pos_tags = []
49 | self.stokens = []
50 | self.emotions_score = []
51 | self.emoticons = []
52 | self.debug = debug
53 |
54 | def apply_emotions(self):
55 | """
56 | Compute emotion scores based on emoticon list.
57 | """
58 | score = 0
59 | for token in self.stokens:
60 | if self.lexicon.has_key(token) \
61 | and self.lexicon[token].has_key('emoticon'):
62 | if self.lexicon[token]['priorpolarity'] == "negative":
63 | score = -2
64 | else:
65 | score = 2
66 | self.emoticons.append(token)
67 | self.emotions_score.append(score)
68 |
69 | def apply_weights(self):
70 | """
71 | Adjust (2*n times) scores for polar expressions.
72 | """
73 | # Strong emotion heuristic based on punctuation
74 | strong_emotion = 1
75 | if self.sentence.endswith("!"):
76 | strong_emotion = 1.5
77 | elif self.sentence.endswith("?"):
78 | strong_emotion = 0.5
79 |
80 | for i,polar in enumerate(self.polar_expressions):
81 | #first rule: if strong double score
82 | if polar in self.strong_polar_expressions:
83 | self.polar_with_score[polar] *= 2
84 | #second rule: if intensified polar double score
85 | if self.intensified_polar(polar):
86 | self.polar_with_score[polar] *= 2
87 | #third rule: if polar expr is adjective double score
88 | if self.polar_with_tags[polar] == "adj":
89 | self.polar_with_score[polar] *= 2
90 |
91 | if i + 1 == len(self.polar_expressions):
92 | self.polar_with_score[polar] *= strong_emotion
93 |
94 | def check_precedings(self, polar, words):
95 | """
96 | Find the indexes of polar words and 5 preceding of it.
97 | """
98 | if words.index(polar) >= 6:
99 | return True
100 | else:
101 | return False
102 |
103 | def classify(self, sentence):
104 | """
105 | Sum up the contextual scores of polar expressions and classify the sentence.
106 | """
107 | self.sentence = sentence
108 | # Extracting Features from sentence
109 | self.extract_features()
110 | # Performing Word Sense Disambiguation
111 | # Extracting Polar Expressions
112 | self.word_sense_disambiguation()
113 | # Checking for negation words
114 | self.negation_modeling()
115 | # Adjusting weights to Polar Expressions
116 | self.apply_weights()
117 | self.apply_emotions()
118 | # Performing polarity classification
119 | [prediction, score] = self.predict_class()
120 |
121 |
122 | if self.debug:
123 | print "\n[*] --------------------RESULTS----------------------"
124 | print Tcolors.ADD + " FEATURE WORDS:", self.feature_words
125 | print Tcolors.ADD + " POLAR EXPRESSIONS FOUND:", self.polar_expressions
126 | print Tcolors.ADD + " POLAR WEIGHTS:", self.polar_with_score
127 | print Tcolors.ADD + " EMOTICONS:", self.emoticons
128 | print Tcolors.ADD + " EMOTION WEIGHTS:", self.emotions_score
129 | print Tcolors.ADD + " PREDICTION: ", prediction, ", WITH CONFIDENCE: ", score
130 | print Tcolors.ADD + " NORMALIZED CONFIDENCE: ", score/len(self.words)
131 |
132 | self.words = [w for w in self.words if w != '']
133 | return prediction, score, score/len(self.words) #normalizedScore
134 |
135 | def extract_features(self):
136 | """
137 | Match positive and negative words of a sentence with a score of +1 or -1
138 | respectively, if found in the lexicon.
139 | """
140 | self.words = self.tokenize_words(self.sentence)
141 | words = self.words
142 | for word in words:
143 | word = word.lower()
144 | if word in self.lexicon:
145 | if self.lexicon[word]['priorpolarity'] == "positive":
146 | if not self.feature_words.has_key(word):
147 | self.feature_words[word] = 1
148 | else:
149 | self.feature_words[word] += 1
150 | elif self.lexicon[word]['priorpolarity'] == "negative":
151 | if not self.feature_words.has_key(word):
152 | self.feature_words[word] = -1
153 | else:
154 | self.feature_words[word] -= 1
155 | else:
156 | self.feature_words[word] = 0
157 |
158 | def intensified_polar(self, polar):
159 | if self.words.index(polar) > 0:
160 | previous_word = self.words[self.words.index(polar) - 1]
161 |
162 | if self.lexicon.has_key(previous_word) and self.lexicon[previous_word]["type"]=="strongsubj" \
163 | and 'adj' in self.lexicon[previous_word]["pos1"]:
164 | return True
165 | return False
166 |
167 | def match_tags(self, pos_tag):
168 | if pos_tag:
169 | if pos_tag.startswith("VB"):
170 | return "verb"
171 | elif pos_tag.startswith("JJ"):
172 | return "adj"
173 | elif pos_tag.startswith("NN"):
174 | return "noun"
175 | elif pos_tag.startswith("RB"):
176 | return "adverb"
177 | else:
178 | return 'anypos'
179 | else:
180 | return "anypos"
181 |
182 | def negation_modeling(self):
183 | """
184 | Examine negation words and reassign polarity.
185 | """
186 | for polar in self.polar_expressions:
187 | has_precedings = self.check_precedings(polar, self.words)
188 | for neg in self.negation_words:
189 | if has_precedings:
190 | if (neg in self.words[(self.words.index(polar)-5):self.words.index(polar)])\
191 | or self.polarity_shifting(polar, self.words[(self.words.index(polar)-5):self.words.index(polar)]):
192 | self.polar_with_score[polar] = self.polar_with_score[polar]*(-1)
193 | break
194 | else:
195 | if neg in self.words[0:self.words.index(polar)] or \
196 | self.polarity_shifting(polar, self.words[0:self.words.index(polar)]):
197 | self.polar_with_score[polar] = self.polar_with_score[polar]*(-1)
198 |
199 | def predict_class(self):
200 | summary = 0
201 | summary = sum([value for value in self.polar_with_score.values()])
202 | summary += sum(self.emotions_score)
203 |
204 | if summary > 0:
205 | return "positive", summary
206 | elif summary < 0:
207 | return "negative", summary
208 | else:
209 | return "neutral", summary
210 |
211 | def polarity_shifting(self, polar, words):
212 | for word in words:
213 | if (self.lexicon.has_key(word) and self.lexicon[word]["type"]=="strongsubj" \
214 | and self.lexicon[word]["priorpolarity"] != self.lexicon[polar]["priorpolarity"] \
215 | and self.lexicon[word]["priorpolarity"]!="neutral") or \
216 | word.endswith("n't"):
217 | # reverse polarity of polar
218 | if self.debug:
219 | print "[!] NEGATION WORD FOUND: ",word
220 | return True
221 |
222 | return False
223 |
224 | def tokenize_words(self, sentence):
225 | words = nltk.word_tokenize(sentence.lower())
226 | self.stokens = sentence.split()
227 | for i,word in enumerate(words):
228 | if word[len(word)-1] in string.punctuation:
229 | words[words.index(word)] = word[0:(len(word)-1)]
230 |
231 | self.words_pos_tags = self.tagger.tag(sentence)
232 | for (word, TAG) in self.words_pos_tags:
233 | word = word.lower()
234 | # Handle punctuation in word_tokenization
235 | if TAG and TAG.startswith("VB"):
236 | if word in words:
237 | words[words.index(word)] = stem(word)
238 | return words
239 |
240 | def word_sense_disambiguation(self):
241 | """
242 | Disambiguate words in a sentence if only if the POS tag of the word matches
243 | the POS tag in the lexicon
244 | """
245 | for (word, TAG) in self.words_pos_tags:
246 | # Handle punctuation in word_tokenization
247 | if word[len(word)-1] in string.punctuation:
248 | word = word[0:(len(word)-1)]
249 | matched_tag = self.match_tags(TAG)
250 | word = word.lower()
251 | words = [word, stem(word)]
252 | for w in words:
253 | if self.feature_words.has_key(w) and (matched_tag in self.lexicon[w]["pos1"] \
254 | or self.lexicon[w]["pos1"][0] == "anypos" \
255 | or matched_tag == "anypos"):
256 | self.polar_expressions.append(w)
257 | self.polar_with_score[w] = self.feature_words[w]
258 | self.polar_with_tags[w] = matched_tag
259 | if self.lexicon[w]['type'] == "strongsubj":
260 | self.strong_polar_expressions.append(w)
261 |
262 | self.polar_expressions = list(set(self.polar_expressions))
263 |
264 |
265 | if __name__ == '__main__':
266 | polarity = PolarityClassifier(sys.argv[1])
267 | polarity.classify()
268 |
269 |
270 |
271 |
272 |
--------------------------------------------------------------------------------
/pos.py:
--------------------------------------------------------------------------------
1 | ####################################################################
2 | # Licence: Creative Commons (see COPYRIGHT) #
3 | # Authors: Nikolaos Pappas, Georgios Katsimpras #
4 | # {nik0spapp, gkatsimpras}@gmail.com #
5 | # Supervisor: Efstathios stamatatos #
6 | # stamatatos@aegean.gr #
7 | # University of the Aegean #
8 | # Department of Information and Communication Systems Engineering #
9 | # Information Management Track (MSc) #
10 | # Karlovasi, Samos #
11 | # Greece #
12 | ####################################################################
13 |
14 | import os
15 | import sys
16 | import pickle
17 | import nltk.corpus, nltk.tag, itertools
18 | from terminal_colors import Tcolors
19 |
20 | class SequentialTagger:
21 | """
22 | Sequential tagger: It uses a sequential tagging method for tagging
23 | untagged sentences. Three tag classifiers are used in sequential
24 | order (Unigram, Bigram and Trigram) that are trained with brown
25 | corpus. Experiments have been made to select this specific UBT
26 | sequence that seems to have better precision than the other
27 | combinations.
28 | """
29 | def __init__(self):
30 |
31 | self.filename = "stored/ubt_tagger.classifier"
32 | try:
33 | self.ubt_tagger = pickle.load(open(self.filename))
34 | print Tcolors.ADD + Tcolors.OKBLUE + " Loaded existing UBT tagger!" + Tcolors.ENDC
35 | except:
36 | print Tcolors.ACT + Tcolors.RED + " Existing UBT tagger not found." + Tcolors.ENDC
37 | print "Path:", "stored/ubt_tagger.classifier"
38 | print "Training..."
39 | brown_review_all = nltk.corpus.brown.tagged_sents()
40 | brown_review_sents = nltk.corpus.brown.tagged_sents(categories=['reviews'])
41 | brown_lore_sents = nltk.corpus.brown.tagged_sents(categories=['lore'])
42 | brown_romance_sents = nltk.corpus.brown.tagged_sents(categories=['romance'])
43 |
44 | brown_train = list(itertools.chain(brown_review_sents[:1000],
45 | brown_lore_sents[:1000],
46 | brown_romance_sents[:1000]))
47 | brown_test = list(itertools.chain(brown_review_sents[1000:2000],
48 | brown_lore_sents[1000:2000],
49 | brown_romance_sents[1000:2000]))
50 |
51 | conll_sents = nltk.corpus.conll2000.tagged_sents()
52 | conll2_sents = nltk.corpus.conll2002.tagged_sents()
53 | conll_train = list(conll_sents[:4000])
54 | conll_test = list(conll_sents[4000:8000])
55 |
56 | treebank_sents = nltk.corpus.treebank.tagged_sents()
57 | treebank_train = list(treebank_sents[:1500])
58 | treebank_test = list(treebank_sents[1500:3000])
59 | train_sents = conll_sents + conll2_sents + treebank_sents + brown_train
60 | test_sents = conll_test
61 |
62 | ubt_tagger = self.backoff_tagger(train_sents, [nltk.tag.UnigramTagger,
63 | nltk.tag.BigramTagger,
64 | nltk.tag.TrigramTagger])
65 | self.ubt_tagger = ubt_tagger
66 | output = open(self.filename,'wb')
67 | pickle.dump(self.ubt_tagger,output)
68 | output.close()
69 |
70 | def backoff_tagger(self, tagged_sents, tagger_classes, backoff=None):
71 | """
72 | Creation the sequential tagger
73 | """
74 | if not backoff:
75 | backoff = tagger_classes[0](tagged_sents)
76 | del tagger_classes[0]
77 |
78 | for cls in tagger_classes:
79 | tagger = cls(tagged_sents, backoff=backoff)
80 | backoff = tagger
81 |
82 | return backoff
83 |
84 | def tag(self, sentence):
85 | """
86 | Method for tagging untagged sentences.
87 | """
88 | words = nltk.word_tokenize(sentence)
89 | return self.ubt_tagger.tag(words)
90 |
--------------------------------------------------------------------------------
/replacer.py:
--------------------------------------------------------------------------------
1 | ####################################################################
2 | # Licence: Creative Commons (see COPYRIGHT) #
3 | # Authors: Nikolaos Pappas, Georgios Katsimpras #
4 | # {nik0spapp, gkatsimpras}@gmail.com #
5 | # Supervisor: Efstathios stamatatos #
6 | # stamatatos@aegean.gr #
7 | # University of the Aegean #
8 | # Department of Information and Communication Systems Engineering #
9 | # Information Management Track (MSc) #
10 | # Karlovasi, Samos #
11 | # Greece #
12 | ####################################################################
13 |
14 | import re
15 | from nltk.corpus import wordnet
16 |
17 | class RepeatReplacer(object):
18 | """
19 | RepeatReplacer: Replaces letters that appear in irregular
20 | repetition inside words.
21 | """
22 |
23 | def __init__(self, lexicon):
24 | self.lexicon = lexicon.words
25 | self.repeat_regexp = re.compile(r'(.*)(.)\2(.*)')
26 | self.repl = r'\1\2\3'
27 |
28 | def replace(self, word):
29 | check = re.sub(r'\!|;|\||\.|\?|,|:|"|\)|\(','',word)
30 | if self.lexicon.has_key(word) and self.lexicon[word].has_key('emoticon'):
31 | return word
32 | if wordnet.synsets(check):
33 | if word == check:
34 | return word
35 | else:
36 | return check + "".join(set(word[len(check):]))
37 | repl_word = self.repeat_regexp.sub(self.repl, word)
38 | if repl_word != word:
39 | return self.replace(repl_word)
40 | else:
41 | return repl_word
42 |
43 |
44 | if __name__ == '__main__':
45 | rr = RepeatReplacer()
46 | example = "sorrryyyyyyyyyyy"
47 | print "Before: " + example
48 | rr.replace(example)
49 | print "After: " + example
50 |
--------------------------------------------------------------------------------
/sentiment.py:
--------------------------------------------------------------------------------
1 | ####################################################################
2 | # Licence: Creative Commons (see COPYRIGHT) #
3 | # Authors: Nikolaos Pappas, Georgios Katsimpras #
4 | # {nik0spapp, gkatsimpras}@gmail.com #
5 | # Supervisor: Efstathios stamatatos #
6 | # stamatatos@aegean.gr #
7 | # University of the Aegean #
8 | # Department of Information and Communication Systems Engineering #
9 | # Information Management Track (MSc) #
10 | # Karlovasi, Samos #
11 | # Greece #
12 | ####################################################################
13 |
14 | import os
15 | import sys
16 | import nltk
17 | import pickle
18 | from bootstrapping import Bootstrapping
19 | from pos import SequentialTagger
20 | from hp_classifiers import HpObj, HpSubj
21 | from polarity import PolarityClassifier
22 | from replacer import RepeatReplacer
23 | from terminal_colors import Tcolors
24 |
25 | DEBUG = False
26 |
27 | class Sentiment:
28 | """
29 | Sentiment: Analyses the global sentiment of given text regions
30 | that are decomposed to sentences, using bootstrapping methods for
31 | subjectivity and polarity classification. All sub modules except
32 | from POS tagging are learning by experience.
33 | """
34 |
35 | def __init__(self):
36 | self.pos_tagger = SequentialTagger()
37 | self.hp_obj = HpObj(debug=DEBUG)
38 | self.hp_subj = HpSubj(debug=DEBUG)
39 | self.lexicon = self.hp_obj.lexicon
40 | self.bootstrapping = Bootstrapping(self.hp_obj, self.hp_subj, self.pos_tagger, debug=DEBUG)
41 | self.sentence_tokenizer = nltk.data.load('tokenizers/punkt/english.pickle')
42 | self.total_sentences = ["good","bad"]
43 | self.total_sentiments = ["positive","negative"]
44 |
45 | def analyze(self, clean_text_areas):
46 | """
47 | Analysis of text regions using the following order: Each sentence per
48 | region is passed from the subjectivity classification using bootstrapping
49 | method and then if it turns out to be subjective it is passed
50 | from the polarity classification using bootstrapping method also.
51 | Finally, it results to a decision for the sentiment of the sentence
52 | and the overall sentiment of the regions.
53 | """
54 | if len(clean_text_areas) > 0:
55 | for clean_text in clean_text_areas:
56 | # Sentence detection
57 | clean_text = self.normalize(clean_text)
58 | try:
59 | sentences = self.sentence_tokenizer.tokenize(clean_text)
60 | except:
61 | return {}
62 | sentiments = []
63 | scores = []
64 | nscores = []
65 | results = {'positive':{'count' : 0, 'score' : 0, 'nscore' : 0},
66 | 'neutral':{'count' : 0, 'score' : 0, 'nscore' : 0},
67 | 'negative':{'count' : 0, 'score' : 0, 'nscore' : 0}}
68 |
69 | print
70 | print Tcolors.ACT + " Checking block of text:"
71 | for i, sentence in enumerate(sentences):
72 | print "[" + str(i+1) + "] " + sentence
73 | for i, sentence in enumerate(sentences):
74 | # Proceed to subjectivity classification (bootstrapping procedure).
75 | # (This step could be skipped in case you deal with subjective sentences only.)
76 | sentiment = ""
77 | previous = ""
78 | next = ""
79 | score = 0
80 | nscore = 0
81 | if i == 0 and i + 1 < len(sentences):
82 | next = sentences[i+1]
83 | elif i != 0 and i < len(sentences):
84 | if i + 1 != len(sentences):
85 | next = sentences[i+1]
86 | previous = sentences[i-1]
87 |
88 | if DEBUG: print Tcolors.ACT + " Analyzing subjectivity..."
89 | result = self.bootstrapping.classify(sentence, previous, next)
90 | if result is None:
91 | res = 'Not found!'
92 | else:
93 | res = result
94 | if DEBUG:
95 | print Tcolors.RES + Tcolors.OKGREEN + " " + res + Tcolors.ENDC
96 | print
97 |
98 | # If sentence is subjective
99 | if result == 'subjective' or result is None:
100 | # Proceed to polarity classification
101 | if DEBUG: print Tcolors.ACT + " Analyzing sentiment..."
102 | polarity_classifier = PolarityClassifier(self.pos_tagger, self.lexicon, debug=DEBUG)
103 | sentiment, score, nscore = polarity_classifier.classify(sentence)
104 | if DEBUG: print Tcolors.RES + Tcolors.OKGREEN + " " + sentiment + Tcolors.ENDC
105 | # If sentence is objective
106 | elif result == 'objective':
107 | sentiment = 'neutral'
108 |
109 | # Collect high-confidence training instances for SVM classifier.
110 | # After the training, SVM can be used to classify new sentences.
111 | #if sentiment != "neutral" and sentiment != "":
112 | #if sentiment != "neutral" and abs(nscore) >= 0.4:
113 | # self.total_sentences.append(sentence)
114 | # self.total_sentiments.append(sentiment)
115 |
116 | # Store results to memory
117 | sentiments.append(sentiment)
118 | scores.append(score)
119 | nscores.append(nscore)
120 |
121 | # Update score
122 | if results.has_key(sentiment):
123 | results[sentiment]['nscore'] += nscore
124 | results[sentiment]['score'] += score
125 | results[sentiment]['count'] += 1
126 |
127 | print
128 | print Tcolors.ACT + " Overall sentiment analysis:"
129 | print Tcolors.BGH
130 | print " Parts: ", len(sentences)
131 | print " Sentiments: ", sentiments
132 | print " Scores: ", scores
133 | print " Results: ", "},\n\t ".join((str)(results).split("}, "))
134 | print Tcolors.C
135 |
136 | pcount = results['positive']['count']
137 | ncount = results['negative']['count']
138 | total = len(sentences)
139 | print Tcolors.BG
140 | print " subjective".ljust(16,"-") + "> %.2f" % ((float)(pcount + ncount)*100 / total) + "%"
141 | print " objective".ljust(16,"-") + "> %.2f" % (100 - ((float)(pcount + ncount)*100 / total)) + "%"
142 | print Tcolors.C
143 | print Tcolors.BGGRAY
144 | for sense in results.keys():
145 | count = results[sense]['count']
146 | percentage = (float)(count) * 100 / (len(sentences))
147 | print " " +sense.ljust(15,"-")+"> %.2f" % (percentage) + "%"
148 |
149 | print Tcolors.C
150 | ssum = sum(scores)
151 | confidence = " (%.2f, %.2f)" % (ssum,sum(nscores))
152 | final_sent = ""
153 | pos = True
154 | if results["negative"]["count"] > len(sentences)*1.0/3:
155 | pos = False
156 |
157 | # Print total sentiment score and normalized sentiment score
158 | if ssum > 0 and pos:
159 | print Tcolors.RES + Tcolors.OKGREEN + " positive" + confidence + Tcolors.C
160 | final_sent = "positive"
161 | elif ssum == 0:
162 | print Tcolors.RES + Tcolors.OKGREEN + " neutral" + confidence + Tcolors.C
163 | final_sent = "neutral"
164 | else:
165 | print Tcolors.RES + Tcolors.OKGREEN + " negative" + confidence + Tcolors.C
166 | final_sent = "negative"
167 | print Tcolors.C
168 |
169 | # Store results
170 | total_result_hash = {'sentences' : sentences,
171 | 'sentiments': sentiments,
172 | 'scores' : scores,
173 | 'nscores' : nscores,
174 | 'results' : results,
175 | 'final' : {final_sent:{'score':ssum,'nscore':sum(nscores)}}}
176 | # Train SVM classifier
177 | # self.train_svm()
178 | return total_result_hash
179 |
180 | def normalize(self, text):
181 | """
182 | Make some word improvements before feeding to the sentence tokenizer.
183 | """
184 | rr = RepeatReplacer(self.lexicon)
185 | normalized_text = []
186 | final = None
187 | try:
188 | for word in text.split():
189 | normal = rr.replace(word.lower())
190 | if word[0].isupper():
191 | normal = normal[0].upper() + normal[1:]
192 |
193 | normalized_text.append(normal)
194 | final = " ".join(normalized_text)
195 | except:
196 | final = text
197 |
198 | return final
199 |
200 | def train_svm(self):
201 | """
202 | Train SVM and store data with pickle.
203 | """
204 | self.svm.train(self.total_sentences, self.total_sentiments)
205 | t_output = open(self.svm_train_filename,'wb')
206 | l_output = open(self.svm_label_filename,'wb')
207 | pickle.dump(self.total_sentences,t_output)
208 | pickle.dump(self.total_sentiments,l_output)
209 | t_output.close()
210 | l_output.close()
211 |
212 |
213 | if __name__ == '__main__':
214 | sentiment = Sentiment()
215 | if len(sys.argv) > 1:
216 | sentiment.analyze([sys.argv[1]])
217 | else:
218 | sentiment.analyze([u"I was blown away by some of the comments here posted by people who is either uneducated, ignorant, self-righteous or all-of-the-above...I'm irritated and saddened as I read these \"finger-pointing\" or \"I'm right and you're wrong\" type of posts! Grow up folks! You're not in grade school...learn to embrace what is positive and move forward to do what is right... I have to give much love and respect to Ronny...your work is AMAZING!!! You cannot fathom how good I feel after I watched this video...regardless of history, politics, or whatever forces that makes what the mid-east today...for what you did and many of the followers in Iran and Palestine ...I BELIEVE TOMORROW WILL BE BETTER!!!!!! My name is Christopher Lee, I'm a nurse in Los Angeles and I {HEART} YOU ALL (especially to all of you beautiful and sweet ladies across the way)!!!!!"])
219 |
--------------------------------------------------------------------------------
/stored/learned_patterns:
--------------------------------------------------------------------------------
1 | (dp0
2 | S'learning '
3 | p1
4 | (dp2
5 | S'type'
6 | p3
7 | S'dobj'
8 | p4
9 | sS'freq'
10 | p5
11 | I8
12 | sS'subj_freq'
13 | p6
14 | I8
15 | sS'display'
16 | p7
17 | S'learning'
18 | p8
19 | sS'prob'
20 | p9
21 | F1.0
22 | ssVwatched
23 | p10
24 | (dp11
25 | S'subj_freq'
26 | p12
27 | I25
28 | sS'freq'
29 | p13
30 | I25
31 | sS'type'
32 | p14
33 | S'dobj'
34 | p15
35 | sS'display'
36 | p16
37 | Vwatched
38 | p17
39 | sS'prob'
40 | p18
41 | F1.0
42 | ssV tracks
43 | p19
44 | (dp20
45 | S'type'
46 | p21
47 | S'subj'
48 | p22
49 | sS'freq'
50 | p23
51 | I9
52 | sS'subj_freq'
53 | p24
54 | I9
55 | sS'display'
56 | p25
57 | Vtracks
58 | p26
59 | sS'prob'
60 | p27
61 | F1.0
62 | ssVimagined
63 | p28
64 | (dp29
65 | g21
66 | S'dobj'
67 | p30
68 | sg23
69 | I9
70 | sg24
71 | I9
72 | sg25
73 | Vimagined
74 | p31
75 | sg27
76 | F1.0
77 | ssVbring
78 | p32
79 | (dp33
80 | g21
81 | g30
82 | sg23
83 | I9
84 | sg24
85 | I9
86 | sg25
87 | Vbring
88 | p34
89 | sg27
90 | F1.0
91 | ssVfeatured
92 | p35
93 | (dp36
94 | g21
95 | g30
96 | sg23
97 | I9
98 | sg24
99 | I9
100 | sg25
101 | Vfeatured
102 | p37
103 | sg27
104 | F1.0
105 | ssV posted people
106 | p38
107 | (dp39
108 | g12
109 | I34
110 | sg13
111 | I34
112 | sg14
113 | S'subj'
114 | p40
115 | sg16
116 | Vposted people
117 | p41
118 | sg18
119 | F1.0
120 | ssV get seat
121 | p42
122 | (dp43
123 | g21
124 | g22
125 | sg23
126 | I9
127 | sg24
128 | I9
129 | sg25
130 | Vget seat
131 | p44
132 | sg27
133 | F1.0
134 | ssS" 'm"
135 | p45
136 | (dp46
137 | S'subj_freq'
138 | p47
139 | I7
140 | sS'freq'
141 | p48
142 | I7
143 | sS'type'
144 | p49
145 | S'subj'
146 | p50
147 | sS'display'
148 | p51
149 | S"'m"
150 | p52
151 | sS'prob'
152 | p53
153 | F1.0
154 | ssVdid
155 | p54
156 | (dp55
157 | g12
158 | I25
159 | sg13
160 | I25
161 | sg14
162 | g15
163 | sg16
164 | Vdid
165 | p56
166 | sg18
167 | F1.0
168 | ssVkeep
169 | p57
170 | (dp58
171 | g21
172 | g30
173 | sg23
174 | I9
175 | sg24
176 | I9
177 | sg25
178 | Vkeep
179 | p59
180 | sg27
181 | F1.0
182 | ssV featured
183 | p60
184 | (dp61
185 | g21
186 | g22
187 | sg23
188 | I9
189 | sg24
190 | I9
191 | sg25
192 | Vfeatured
193 | p62
194 | sg27
195 | F1.0
196 | ssS' found'
197 | p63
198 | (dp64
199 | S'subj_freq'
200 | p65
201 | I4
202 | sS'freq'
203 | p66
204 | I4
205 | sS'type'
206 | p67
207 | S'subj'
208 | p68
209 | sS'display'
210 | p69
211 | S'found'
212 | p70
213 | sS'prob'
214 | p71
215 | F1.0
216 | ssVconveyed
217 | p72
218 | (dp73
219 | g21
220 | g30
221 | sg23
222 | I9
223 | sg24
224 | I9
225 | sg25
226 | Vconveyed
227 | p74
228 | sg27
229 | F1.0
230 | ssVCalled
231 | p75
232 | (dp76
233 | g21
234 | g30
235 | sg23
236 | I23
237 | sg24
238 | I23
239 | sg25
240 | VCalled
241 | p77
242 | sg27
243 | F1.0
244 | ssVwait
245 | p78
246 | (dp79
247 | g21
248 | g30
249 | sg23
250 | I7
251 | sg24
252 | I7
253 | sg25
254 | Vwait
255 | p80
256 | sg27
257 | F1.0
258 | ssVforces
259 | p81
260 | (dp82
261 | g12
262 | I25
263 | sg13
264 | I25
265 | sg14
266 | g15
267 | sg16
268 | Vforces
269 | p83
270 | sg18
271 | F1.0
272 | ssV did
273 | p84
274 | (dp85
275 | g12
276 | I25
277 | sg13
278 | I25
279 | sg14
280 | g40
281 | sg16
282 | Vdid
283 | p86
284 | sg18
285 | F1.0
286 | ssV blown
287 | p87
288 | (dp88
289 | g12
290 | I34
291 | sg13
292 | I34
293 | sg14
294 | g40
295 | sg16
296 | Vblown
297 | p89
298 | sg18
299 | F1.0
300 | ssS"'m "
301 | p90
302 | (dp91
303 | g47
304 | I7
305 | sg48
306 | I7
307 | sg49
308 | S'dobj'
309 | p92
310 | sg51
311 | S"'m"
312 | p93
313 | sg53
314 | F1.0
315 | ssV keep
316 | p94
317 | (dp95
318 | g21
319 | g22
320 | sg23
321 | I9
322 | sg24
323 | I9
324 | sg25
325 | Vkeep
326 | p96
327 | sg27
328 | F1.0
329 | ssVposted
330 | p97
331 | (dp98
332 | g12
333 | I34
334 | sg13
335 | I34
336 | sg14
337 | g15
338 | sg16
339 | Vposted
340 | p99
341 | sg18
342 | F1.0
343 | ssV Called L
344 | p100
345 | (dp101
346 | g21
347 | g22
348 | sg23
349 | I7
350 | sg24
351 | I7
352 | sg25
353 | VCalled L
354 | p102
355 | sg27
356 | F1.0
357 | ssVreleased
358 | p103
359 | (dp104
360 | g21
361 | g30
362 | sg23
363 | I9
364 | sg24
365 | I9
366 | sg25
367 | Vreleased
368 | p105
369 | sg27
370 | F1.0
371 | ssS'found '
372 | p106
373 | (dp107
374 | g65
375 | I4
376 | sg66
377 | I4
378 | sg67
379 | S'dobj'
380 | p108
381 | sg69
382 | S'found'
383 | p109
384 | sg71
385 | F1.0
386 | ssVblown
387 | p110
388 | (dp111
389 | g12
390 | I34
391 | sg13
392 | I34
393 | sg14
394 | g15
395 | sg16
396 | Vblown
397 | p112
398 | sg18
399 | F1.0
400 | ssVCompiled
401 | p113
402 | (dp114
403 | g21
404 | g30
405 | sg23
406 | I9
407 | sg24
408 | I9
409 | sg25
410 | VCompiled
411 | p115
412 | sg27
413 | F1.0
414 | ssV get
415 | p116
416 | (dp117
417 | g21
418 | g22
419 | sg23
420 | I9
421 | sg24
422 | I9
423 | sg25
424 | Vget
425 | p118
426 | sg27
427 | F1.0
428 | ssV give
429 | p119
430 | (dp120
431 | g12
432 | I37
433 | sg13
434 | I37
435 | sg14
436 | g40
437 | sg16
438 | Vgive
439 | p121
440 | sg18
441 | F1.0
442 | ssV conveyed
443 | p122
444 | (dp123
445 | g21
446 | g22
447 | sg23
448 | I9
449 | sg24
450 | I9
451 | sg25
452 | Vconveyed
453 | p124
454 | sg27
455 | F1.0
456 | ssV forces
457 | p125
458 | (dp126
459 | g12
460 | I25
461 | sg13
462 | I25
463 | sg14
464 | g40
465 | sg16
466 | Vforces
467 | p127
468 | sg18
469 | F1.0
470 | ssV released
471 | p128
472 | (dp129
473 | g21
474 | g22
475 | sg23
476 | I9
477 | sg24
478 | I9
479 | sg25
480 | Vreleased
481 | p130
482 | sg27
483 | F1.0
484 | ssV was blown
485 | p131
486 | (dp132
487 | g12
488 | I34
489 | sg13
490 | I34
491 | sg14
492 | g40
493 | sg16
494 | Vwas blown
495 | p133
496 | sg18
497 | F1.0
498 | ssV capture
499 | p134
500 | (dp135
501 | g21
502 | g22
503 | sg23
504 | I9
505 | sg24
506 | I9
507 | sg25
508 | Vcapture
509 | p136
510 | sg27
511 | F1.0
512 | ssVtracks
513 | p137
514 | (dp138
515 | g21
516 | g30
517 | sg23
518 | I9
519 | sg24
520 | I9
521 | sg25
522 | Vtracks
523 | p139
524 | sg27
525 | F1.0
526 | ssS' found presentation'
527 | p140
528 | (dp141
529 | g65
530 | I4
531 | sg66
532 | I4
533 | sg67
534 | g68
535 | sg69
536 | S'found presentation'
537 | p142
538 | sg71
539 | F1.0
540 | ssS' think'
541 | p143
542 | (dp144
543 | S'type'
544 | p145
545 | S'subj'
546 | p146
547 | sS'freq'
548 | p147
549 | I4
550 | sS'subj_freq'
551 | p148
552 | I4
553 | sS'display'
554 | p149
555 | S'think'
556 | p150
557 | sS'prob'
558 | p151
559 | F1.0
560 | ssVgive
561 | p152
562 | (dp153
563 | g12
564 | I37
565 | sg13
566 | I37
567 | sg14
568 | g15
569 | sg16
570 | Vgive
571 | p154
572 | sg18
573 | F1.0
574 | ssS'think '
575 | p155
576 | (dp156
577 | g145
578 | S'dobj'
579 | p157
580 | sg147
581 | I4
582 | sg148
583 | I4
584 | sg149
585 | S'think'
586 | p158
587 | sg151
588 | F1.0
589 | ssV posted
590 | p159
591 | (dp160
592 | g12
593 | I34
594 | sg13
595 | I34
596 | sg14
597 | g40
598 | sg16
599 | Vposted
600 | p161
601 | sg18
602 | F1.0
603 | ssV Called D
604 | p162
605 | (dp163
606 | g21
607 | g22
608 | sg23
609 | I7
610 | sg24
611 | I7
612 | sg25
613 | VCalled D
614 | p164
615 | sg27
616 | F1.0
617 | ssV Called
618 | p165
619 | (dp166
620 | g21
621 | g22
622 | sg23
623 | I23
624 | sg24
625 | I23
626 | sg25
627 | VCalled
628 | p167
629 | sg27
630 | F1.0
631 | ssS'sharing '
632 | p168
633 | (dp169
634 | g21
635 | g30
636 | sg23
637 | I14
638 | sg24
639 | I0
640 | sg25
641 | S'sharing'
642 | p170
643 | sg27
644 | F0.0
645 | ssS' learning'
646 | p171
647 | (dp172
648 | g3
649 | S'subj'
650 | p173
651 | sg5
652 | I8
653 | sg6
654 | I8
655 | sg7
656 | S'learning'
657 | p174
658 | sg9
659 | F1.0
660 | ssVget
661 | p175
662 | (dp176
663 | g21
664 | g30
665 | sg23
666 | I9
667 | sg24
668 | I9
669 | sg25
670 | Vget
671 | p177
672 | sg27
673 | F1.0
674 | ssV wait
675 | p178
676 | (dp179
677 | g21
678 | g22
679 | sg23
680 | I7
681 | sg24
682 | I7
683 | sg25
684 | Vwait
685 | p180
686 | sg27
687 | F1.0
688 | ssV Called Ad
689 | p181
690 | (dp182
691 | g21
692 | g22
693 | sg23
694 | I14
695 | sg24
696 | I14
697 | sg25
698 | VCalled Ad
699 | p183
700 | sg27
701 | F1.0
702 | ssVread
703 | p184
704 | (dp185
705 | g12
706 | I34
707 | sg13
708 | I34
709 | sg14
710 | g15
711 | sg16
712 | Vread
713 | p186
714 | sg18
715 | F1.0
716 | ssV imagined
717 | p187
718 | (dp188
719 | g21
720 | g22
721 | sg23
722 | I9
723 | sg24
724 | I9
725 | sg25
726 | Vimagined
727 | p189
728 | sg27
729 | F1.0
730 | ssV watched
731 | p190
732 | (dp191
733 | g12
734 | I25
735 | sg13
736 | I25
737 | sg14
738 | g40
739 | sg16
740 | Vwatched
741 | p192
742 | sg18
743 | F1.0
744 | ssS' sharing'
745 | p193
746 | (dp194
747 | g21
748 | g22
749 | sg23
750 | I14
751 | sg24
752 | I0
753 | sg25
754 | S'sharing'
755 | p195
756 | sg27
757 | F0.0
758 | ssV bring
759 | p196
760 | (dp197
761 | g21
762 | g22
763 | sg23
764 | I9
765 | sg24
766 | I9
767 | sg25
768 | Vbring
769 | p198
770 | sg27
771 | F1.0
772 | ssVcapture
773 | p199
774 | (dp200
775 | g21
776 | g30
777 | sg23
778 | I9
779 | sg24
780 | I9
781 | sg25
782 | Vcapture
783 | p201
784 | sg27
785 | F1.0
786 | ssV Compiled
787 | p202
788 | (dp203
789 | g21
790 | g22
791 | sg23
792 | I9
793 | sg24
794 | I9
795 | sg25
796 | VCompiled
797 | p204
798 | sg27
799 | F1.0
800 | ssV imagined worlds
801 | p205
802 | (dp206
803 | g21
804 | g22
805 | sg23
806 | I9
807 | sg24
808 | I9
809 | sg25
810 | Vimagined worlds
811 | p207
812 | sg27
813 | F1.0
814 | ssV Called Land
815 | p208
816 | (dp209
817 | g21
818 | g22
819 | sg23
820 | I23
821 | sg24
822 | I23
823 | sg25
824 | VCalled Land
825 | p210
826 | sg27
827 | F1.0
828 | ssV read
829 | p211
830 | (dp212
831 | g12
832 | I34
833 | sg13
834 | I34
835 | sg14
836 | g40
837 | sg16
838 | Vread
839 | p213
840 | sg18
841 | F1.0
842 | ss.
--------------------------------------------------------------------------------
/stored/ubt_tagger.classifier:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nik0spapp/usent/52fe720964001344897db2ad9a98cb53a64be717/stored/ubt_tagger.classifier
--------------------------------------------------------------------------------
/svm.py:
--------------------------------------------------------------------------------
1 | ####################################################################
2 | # Licence: Creative Commons (see COPYRIGHT) #
3 | # Authors: Nikolaos Pappas, Georgios Katsimpras #
4 | # {nik0spapp, gkatsimpras}@gmail.com #
5 | # Supervisor: Efstathios stamatatos #
6 | # stamatatos@aegean.gr #
7 | # University of the Aegean #
8 | # Department of Information and Communication Systems Engineering #
9 | # Information Management Track (MSc) #
10 | # Karlovasi, Samos #
11 | # Greece #
12 | ####################################################################
13 |
14 | import nltk
15 | import sys
16 | import os
17 | import pickle
18 | import numpy as np
19 | from PyML import svm, ker, featsel
20 | from PyML.containers.vectorDatasets import SparseDataSet, VectorDataSet
21 | from PyML.classifiers.composite import Chain, FeatureSelect
22 | from scrapy.conf import settings
23 | from terminal_colors import Tcolors
24 | from PyML.classifiers.svm import loadSVM
25 |
26 | class SvmClassifier:
27 | """
28 | SVM classifier: Performing training and prediction of sentiment class.
29 | """
30 | def __init__(self, lexicon, C=1, num_features=100):
31 | self.training_set = None
32 | self.classes = None
33 | self.test_set = None
34 | self.results = None
35 | self.kernel = ker.Linear()
36 | self.C = C
37 | self.feature_data = PATH + "/learning/stored/feature.data"
38 | self.label_data = PATH + "/learning/stored/svm_label.data"
39 | self.lexicon = lexicon
40 | self.num_features = len(self.lexicon.words.keys())
41 | try:
42 | print "Loading existing SVM..."
43 | features = pickle.load(open(self.feature_data))
44 | labels = pickle.load(open(self.label_data))
45 | sparsedata = SparseDataSet(features, L=labels)
46 | self.svm_classifier = loadSVM(PATH + "/learning/stored/svm.classifier",sparsedata)
47 | except Exception as e:
48 | print e
49 | print "Existing SVM not found!"
50 | self.svm_classifier = svm.SVM(self.kernel)
51 | self.accuracy = None
52 | self.predicted_labels = None
53 | score = featsel.FeatureScore('golub')
54 | self.filter = featsel.Filter(score)
55 | self.feature_selector = FeatureSelect(self.svm_classifier, self.filter)
56 | self.chain = Chain([self.feature_selector, self.svm_classifier])
57 |
58 | def classify(self, sentences, labels):
59 | self.test_set = self.compute_features(sentences)
60 | print
61 | print Tcolors.ACT + " Classifying instance with SVM: " + Tcolors.RED + sentences[0] + Tcolors.C
62 | print Tcolors.HEADER
63 | test_data = SparseDataSet(self.test_set, L=labels)
64 | self.results = self.svm_classifier.test(test_data)
65 | print Tcolors.C
66 | return self.results
67 |
68 | def compute_features(self, sentences):
69 | features = []
70 | for i,sent in enumerate(sentences):
71 | sent = sent.lower()
72 | words = nltk.word_tokenize(sent)
73 | feature = np.zeros(self.num_features)
74 | for word in words:
75 | if word.lower() in self.lexicon.words.keys():
76 | feature[self.lexicon.words.keys().index(word)] = 1
77 | features.append(feature)
78 | return features
79 |
80 | def initialize_lexicon(self):
81 | pass
82 |
83 | def print_stats(self):
84 | print "[*] SVM Classifier ACCURACY: ", self.accuracy
85 | print "[*] SVM Classifier PREDICTED_LABEL: ", self.predicted_labels[0]
86 |
87 | def stats(self):
88 | self.accuracy = self.results.getSuccessRate()
89 | self.predicted_labels = self.results.getPredictedLabels()
90 |
91 | def save(self,data,features,labels):
92 | output = open(self.feature_data ,'wb')
93 | pickle.dump(features,output)
94 | output.close()
95 | output = open(self.label_data,'wb')
96 | pickle.dump(labels,output)
97 | output.close()
98 | self.svm_classifier.save(PATH + "/learning/stored/svm.classifier")
99 |
100 | def train(self, training_set, labels):
101 | print Tcolors.ACT + " Training SVM with chaining..."
102 | features = self.compute_features(training_set)
103 | data = SparseDataSet(features, L=labels)
104 | print Tcolors.CYAN
105 | self.training_set = data
106 | self.svm_classifier.train(data)
107 | self.save(data,features,labels)
108 | print Tcolors.C
109 |
110 |
111 |
--------------------------------------------------------------------------------
/terminal_colors.py:
--------------------------------------------------------------------------------
1 | ####################################################################
2 | # Licence: Creative Commons (see COPYRIGHT) #
3 | # Authors: Nikolaos Pappas, Georgios Katsimpras #
4 | # {nik0spapp, gkatsimpras}@gmail.com #
5 | # Supervisor: Efstathios stamatatos #
6 | # stamatatos@aegean.gr #
7 | # University of the Aegean #
8 | # Department of Information and Communication Systems Engineering #
9 | # Information Management Track (MSc) #
10 | # Karlovasi, Samos #
11 | # Greece #
12 | ####################################################################
13 |
14 | class Tcolors:
15 | HEADER = '\033[1;95m'
16 | OKBLUE = '\033[94m'
17 | OKGREEN = '\033[1;92m'
18 | WARNING = '\033[1;93m'
19 | W = '\033[1;37m'
20 | GRAY = W
21 | BGGRAY = '\033[1;37;40m'
22 | BG = '\033[1;30;47m'
23 | BGH = '\033[1;40;41m'
24 | FAIL = '\033[91m'
25 | RED = '\033[1;91m'
26 | ENDC = '\033[0m'
27 | CYAN = '\033[1;36m'
28 | INF = '\033[1;90m'
29 | C = ENDC
30 |
31 | ACT = W + "["+RED+"*"+ENDC+W+"]" + C
32 | PROC = W + "["+OKBLUE+"*"+ENDC+W+"]" + C
33 | ADD = W + "["+WARNING+"+"+ENDC+W+"]" + C
34 | RES = W + "["+OKGREEN+"x"+ENDC+W+"]" + C
35 | INFO = W + "["+OKBLUE+"INFO:"+ENDC+W+"]" + C
36 | OK = W + "[ "+ OKGREEN + "OK" + ENDC+W+ " ]" + C
37 | def disable(self):
38 | self.HEADER = ''
39 | self.OKBLUE = ''
40 | self.OKGREEN = ''
41 | self.WARNING = ''
42 | self.FAIL = ''
43 | self.ENDC = ''
44 |
--------------------------------------------------------------------------------