├── DataPrep.py
├── FeatureSelection.py
├── LICENSE
├── README.md
├── __pycache__
    ├── DataPrep.cpython-36.pyc
    └── FeatureSelection.cpython-36.pyc
├── _config.yml
├── animate.css
├── bootstrap.css
├── bootstrap.min.css
├── classifier.py
├── final-fnd.ipynb
├── final_model.sav
├── font.css
├── front.py
├── images
    ├── LR_LCurve.PNG
    ├── ProcessFlow.PNG
    └── RF_LCurve.png
├── index
├── li-scroller.css
├── liar_dataset
    ├── README
    ├── test.tsv
    ├── train.tsv
    └── valid.tsv
├── model.pkl
├── my_model
├── prediction.py
├── project.css
├── style.css
├── test.csv
├── theme.css
├── train.csv
└── valid.csv


/DataPrep.py:
--------------------------------------------------------------------------------
  1 | # -*- coding: utf-8 -*-
  2 | """
  3 | Created on Sat Nov  4 12:00:49 2017
  4 | 
  5 | @author: NishitP
  6 | """
  7 | #import os
  8 | import pandas as pd
  9 | import csv
 10 | import numpy as np
 11 | import nltk
 12 | from nltk.stem import SnowballStemmer
 13 | from nltk.stem.porter import PorterStemmer
 14 | from nltk.tokenize import word_tokenize
 15 | import seaborn as sb
 16 | 
 17 | #before reading the files, setup the working directory to point to project repo
 18 | #reading data files 
 19 | 
 20 | 
 21 | test_filename = 'test.csv'
 22 | train_filename = 'train.csv'
 23 | valid_filename = 'valid.csv'
 24 | 
 25 | train_news = pd.read_csv(train_filename)
 26 | test_news = pd.read_csv(test_filename)
 27 | valid_news = pd.read_csv(valid_filename)
 28 | 
 29 | 
 30 | 
 31 | #data observation
 32 | def data_obs():
 33 |     print("training dataset size:")
 34 |     print(train_news.shape)
 35 |     print(train_news.head(10))
 36 | 
 37 |     #below dataset were used for testing and validation purposes
 38 |     print(test_news.shape)
 39 |     print(test_news.head(10))
 40 |     
 41 |     print(valid_news.shape)
 42 |     print(valid_news.head(10))
 43 | 
 44 | #check the data by calling below function
 45 | #data_obs()
 46 | 
 47 | #distribution of classes for prediction
 48 | def create_distribution(dataFile):
 49 |     
 50 |     return sb.countplot(x='Label', data=dataFile, palette='hls')
 51 |     
 52 | 
 53 | #by calling below we can see that training, test and valid data seems to be failry evenly distributed between the classes
 54 | create_distribution(train_news)
 55 | create_distribution(test_news)
 56 | create_distribution(valid_news)
 57 | 
 58 | 
 59 | #data integrity check (missing label values)
 60 | #none of the datasets contains missing values therefore no cleaning required
 61 | def data_qualityCheck():
 62 |     
 63 |     print("Checking data qualitites...")
 64 |     train_news.isnull().sum()
 65 |     train_news.info()
 66 |         
 67 |     print("check finished.")
 68 | 
 69 |     #below datasets were used to 
 70 |     test_news.isnull().sum()
 71 |     test_news.info()
 72 | 
 73 |     valid_news.isnull().sum()
 74 |     valid_news.info()
 75 | 
 76 | #run the below function call to see the quality check results
 77 | #data_qualityCheck()
 78 | 
 79 | 
 80 | 
 81 | #eng_stemmer = SnowballStemmer('english')
 82 | #stopwords = set(nltk.corpus.stopwords.words('english'))
 83 | 
 84 | #Stemming
 85 | def stem_tokens(tokens, stemmer):
 86 |     stemmed = []
 87 |     for token in tokens:
 88 |         stemmed.append(stemmer.stem(token))
 89 |     return stemmed
 90 | 
 91 | #process the data
 92 | def process_data(data,exclude_stopword=True,stem=True):
 93 |     tokens = [w.lower() for w in data]
 94 |     tokens_stemmed = tokens
 95 |     tokens_stemmed = stem_tokens(tokens, eng_stemmer)
 96 |     tokens_stemmed = [w for w in tokens_stemmed if w not in stopwords ]
 97 |     return tokens_stemmed
 98 | 
 99 | 
100 | #creating ngrams
101 | #unigram 
102 | def create_unigram(words):
103 |     assert type(words) == list
104 |     return words
105 | 
106 | #bigram
107 | def create_bigrams(words):
108 |     assert type(words) == list
109 |     skip = 0
110 |     join_str = " "
111 |     Len = len(words)
112 |     if Len > 1:
113 |         lst = []
114 |         for i in range(Len-1):
115 |             for k in range(1,skip+2):
116 |                 if i+k < Len:
117 |                     lst.append(join_str.join([words[i],words[i+k]]))
118 |     else:
119 |         #set it as unigram
120 |         lst = create_unigram(words)
121 |     return lst
122 | 
123 | """
124 | #trigrams
125 | def create_trigrams(words):
126 |     assert type(words) == list
127 |     skip == 0
128 |     join_str = " "
129 |     Len = len(words)
130 |     if L > 2:
131 |         lst = []
132 |         for i in range(1,skip+2):
133 |             for k1 in range(1, skip+2):
134 |                 for k2 in range(1,skip+2):
135 |                     for i+k1 < Len and i+k1+k2 < Len:
136 |                         lst.append(join_str.join([words[i], words[i+k1],words[i+k1+k2])])
137 |         else:
138 |             #set is as bigram
139 |             lst = create_bigram(words)
140 |     return lst
141 | """
142 | 
143 | 
144 | porter = PorterStemmer()
145 | 
146 | def tokenizer(text):
147 |     return text.split()
148 | 
149 | 
150 | def tokenizer_porter(text):
151 |     return [porter.stem(word) for word in text.split()]
152 | 
153 | #doc = ['runners like running and thus they run','this is a test for tokens']
154 | #tokenizer([word for line in test_news.iloc[:,1] for word in line.lower().split()])
155 | 
156 | #show the distribution of labels in the train and test data
157 | """def create_datafile(filename)
158 |     #function to slice the dataframe to keep variables necessary to be used for classification
159 |     return "return df to be used"
160 | """
161 |     
162 | """#converting multiclass labels present in our datasets to binary class labels
163 | for i , row in data_TrainNews.iterrows():
164 |     if (data_TrainNews.iloc[:,0] == "mostly-true" | data_TrainNews.iloc[:,0] == "half-true" | data_TrainNews.iloc[:,0] == "true"):
165 |         data_TrainNews.iloc[:,0] = "true"
166 |     else :
167 |         data_TrainNews.iloc[:,0] = "false"
168 |         
169 | for i,row in data_TrainNews.iterrows():
170 |     print(row)
171 | """
172 |     
173 | 


--------------------------------------------------------------------------------
/FeatureSelection.py:
--------------------------------------------------------------------------------
  1 | # -*- coding: utf-8 -*-
  2 | """
  3 | Created on Sat Nov  4 14:13:38 2017
  4 | 
  5 | @author: NishitP
  6 | 
  7 | Note: before we can train an algorithm to classify fake news labels, we need to extract features from it. It means reducing the mass
  8 | of unstructured data into some uniform set of attributes that an algorithm can understand. For fake news detection, it could be 
  9 | word counts (bag of words). 
 10 | """
 11 | import DataPrep
 12 | import pandas as pd
 13 | import numpy as np
 14 | from sklearn.feature_extraction.text import CountVectorizer
 15 | from sklearn.feature_extraction.text import TfidfTransformer
 16 | from sklearn.feature_extraction.text import TfidfVectorizer
 17 | from sklearn.pipeline import Pipeline
 18 | import nltk
 19 | import nltk.corpus 
 20 | from nltk.tokenize import word_tokenize
 21 | from gensim.models.word2vec import Word2Vec
 22 | 
 23 | 
 24 | #we will start with simple bag of words technique 
 25 | #creating feature vector - document term matrix
 26 | countV = CountVectorizer()
 27 | train_count = countV.fit_transform(DataPrep.train_news['Statement'].values)
 28 | 
 29 | print(countV)
 30 | print(train_count)
 31 | 
 32 | #print training doc term matrix
 33 | #we have matrix of size of (10240, 12196) by calling below
 34 | def get_countVectorizer_stats():
 35 |     
 36 |     #vocab size
 37 |     train_count.shape
 38 | 
 39 |     #check vocabulary using below command
 40 |     print(countV.vocabulary_)
 41 | 
 42 |     #get feature names
 43 |     print(countV.get_feature_names()[:25])
 44 | 
 45 | 
 46 | #create tf-df frequency features
 47 | #tf-idf 
 48 | tfidfV = TfidfTransformer()
 49 | train_tfidf = tfidfV.fit_transform(train_count)
 50 | 
 51 | def get_tfidf_stats():
 52 |     train_tfidf.shape
 53 |     #get train data feature names 
 54 |     print(train_tfidf.A[:10])
 55 | 
 56 | 
 57 | #bag of words - with n-grams
 58 | #countV_ngram = CountVectorizer(ngram_range=(1,3),stop_words='english')
 59 | #tfidf_ngram  = TfidfTransformer(use_idf=True,smooth_idf=True)
 60 | 
 61 | tfidf_ngram = TfidfVectorizer(stop_words='english',ngram_range=(1,4),use_idf=True,smooth_idf=True)
 62 | 
 63 | 
 64 | #POS Tagging
 65 | tagged_sentences = nltk.corpus.treebank.tagged_sents()
 66 | 
 67 | cutoff = int(.75 * len(tagged_sentences))
 68 | training_sentences = DataPrep.train_news['Statement']
 69 |  
 70 | print(training_sentences)
 71 | 
 72 | #training POS tagger based on words
 73 | def features(sentence, index):
 74 |     """ sentence: [w1, w2, ...], index: the index of the word """
 75 |     return {
 76 |         'word': sentence[index],
 77 |         'is_first': index == 0,
 78 |         'is_last': index == len(sentence) - 1,
 79 |         'is_capitalized': sentence[index][0].upper() == sentence[index][0],
 80 |         'is_all_caps': sentence[index].upper() == sentence[index],
 81 |         'is_all_lower': sentence[index].lower() == sentence[index],
 82 |         'prefix-1': sentence[index][0],
 83 |         'prefix-2': sentence[index][:2],
 84 |         'prefix-3': sentence[index][:3],
 85 |         'suffix-1': sentence[index][-1],
 86 |         'suffix-2': sentence[index][-2:],
 87 |         'suffix-3': sentence[index][-3:],
 88 |         'prev_word': '' if index == 0 else sentence[index - 1],
 89 |         'next_word': '' if index == len(sentence) - 1 else sentence[index + 1],
 90 |         'has_hyphen': '-' in sentence[index],
 91 |         'is_numeric': sentence[index].isdigit(),
 92 |         'capitals_inside': sentence[index][1:].lower() != sentence[index][1:]
 93 |     }
 94 |     
 95 |     
 96 | #helper function to strip tags from tagged corpus	
 97 | def untag(tagged_sentence):
 98 |     return [w for w, t in tagged_sentence]
 99 | 
100 | 
101 | 
102 | #Using Word2Vec 
103 | with open("glove.6B.50d.txt", "rb") as lines:
104 |     w2v = {line.split()[0]: np.array(map(float, line.split()[1:]))
105 |            for line in lines}
106 | 
107 | 
108 | 
109 | #model = gensim.models.Word2Vec(X, size=100) # x be tokenized text
110 | #w2v = dict(zip(model.wv.index2word, model.wv.syn0))
111 | 
112 | 
113 | class MeanEmbeddingVectorizer(object):
114 |     def __init__(self, word2vec):
115 |         self.word2vec = word2vec
116 |         # if a text is empty we should return a vector of zeros
117 |         # with the same dimensionality as all the other vectors
118 |         self.dim = len(word2vec.itervalues().next())
119 | 
120 |     def fit(self, X, y):
121 |         return self
122 | 
123 |     def transform(self, X):
124 |         return np.array([
125 |             np.mean([self.word2vec[w] for w in words if w in self.word2vec]
126 |                     or [np.zeros(self.dim)], axis=0)
127 |             for words in X
128 |         ])
129 | 
130 | 
131 | """
132 | class TfidfEmbeddingVectorizer(object):
133 |     def __init__(self, word2vec):
134 |         self.word2vec = word2vec
135 |         self.word2weight = None
136 |         self.dim = len(word2vec.itervalues().next())
137 | 
138 |     def fit(self, X, y):
139 |         tfidf = TfidfVectorizer(analyzer=lambda x: x)
140 |         tfidf.fit(X)
141 |         # if a word was never seen - it must be at least as infrequent
142 |         # as any of the known words - so the default idf is the max of 
143 |         # known idf's
144 |         max_idf = max(tfidf.idf_)
145 |         self.word2weight = defaultdict(
146 |             lambda: max_idf,
147 |             [(w, tfidf.idf_[i]) for w, i in tfidf.vocabulary_.items()])
148 | 
149 |         return self
150 | 
151 |     def transform(self, X):
152 |         return np.array([
153 |                 np.mean([self.word2vec[w] * self.word2weight[w]
154 |                          for w in words if w in self.word2vec] or
155 |                         [np.zeros(self.dim)], axis=0)
156 |                 for words in X
157 |             ])
158 | 
159 | """
160 | 


--------------------------------------------------------------------------------
/LICENSE:
--------------------------------------------------------------------------------
 1 | MIT License
 2 | 
 3 | Copyright (c) 2018 Nishit Patel
 4 | 
 5 | Permission is hereby granted, free of charge, to any person obtaining a copy
 6 | of this software and associated documentation files (the "Software"), to deal
 7 | in the Software without restriction, including without limitation the rights
 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
 9 | copies of the Software, and to permit persons to whom the Software is
10 | furnished to do so, subject to the following conditions:
11 | 
12 | The above copyright notice and this permission notice shall be included in all
13 | copies or substantial portions of the Software.
14 | 
15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
21 | SOFTWARE.
22 | 


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
  1 | # Fake News Detection
  2 | 
  3 | Fake News Detection in Python
  4 | 
  5 | In this project, we have used various natural language processing techniques and machine learning algorithms to classify fake news articles using sci-kit libraries from python. 
  6 | 
  7 | ## Getting Started
  8 | 
  9 | These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.
 10 | 
 11 | ### Prerequisites
 12 | 
 13 | What things you need to install the software and how to install them:
 14 | 
 15 | 1. Python 3.6 
 16 |    - This setup requires that your machine has python 3.6 installed on it. you can refer to this url https://www.python.org/downloads/ to download python. Once you have python downloaded and installed, you will need to setup PATH variables (if you want to run python program directly, detail instructions are below in *how to run software section*). To do that check this: https://www.pythoncentral.io/add-python-to-path-python-is-not-recognized-as-an-internal-or-external-command/.  
 17 |    - Setting up PATH variable is optional as you can also run program without it and more instruction are given below on this topic. 
 18 | 2. Second and easier option is to download anaconda and use its anaconda prompt to run the commands. To install anaconda check this url https://www.anaconda.com/download/
 19 | 3. You will also need to download and install below 3 packages after you install either python or anaconda from the steps above
 20 |    - Sklearn (scikit-learn)
 21 |    - numpy
 22 |    - scipy
 23 |    
 24 |   - if you have chosen to install python 3.6 then run below commands in command prompt/terminal to install these packages
 25 |    ```
 26 |    pip install -U scikit-learn
 27 |    pip install numpy
 28 |    pip install scipy
 29 |    ```
 30 |    - if you have chosen to install anaconda then run below commands in anaconda prompt to install these packages
 31 |    ```
 32 |    conda install -c scikit-learn
 33 |    conda install -c anaconda numpy
 34 |    conda install -c anaconda scipy
 35 |    ```   
 36 | 
 37 | #### Dataset used
 38 | The data source used for this project is LIAR dataset which contains 3 files with .tsv format for test, train and validation. Below is some description about the data files used for this project.
 39 | 	
 40 | LIAR: A BENCHMARK DATASET FOR FAKE NEWS DETECTION
 41 | 
 42 | William Yang Wang, "Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detection, to appear in Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (ACL 2017), short paper, Vancouver, BC, Canada, July 30-August 4, ACL.
 43 | 
 44 | the original dataset contained 13 variables/columns for train, test and validation sets as follows:
 45 | 
 46 | * Column 1: the ID of the statement ([ID].json).
 47 | * Column 2: the label. (Label class contains: True, Mostly-true, Half-true, Barely-true, FALSE, Pants-fire)
 48 | * Column 3: the statement.
 49 | * Column 4: the subject(s).
 50 | * Column 5: the speaker.
 51 | * Column 6: the speaker's job title.
 52 | * Column 7: the state info.
 53 | * Column 8: the party affiliation.
 54 | * Column 9-13: the total credit history count, including the current statement.
 55 | * 9: barely true counts.
 56 | * 10: false counts.
 57 | * 11: half true counts.
 58 | * 12: mostly true counts.
 59 | * 13: pants on fire counts.
 60 | * Column 14: the context (venue / location of the speech or statement).
 61 | 
 62 | To make things simple we have chosen only 2 variables from this original dataset for this classification. The other variables can be added later to add some more complexity and enhance the features.
 63 | 
 64 | Below are the columns used to create 3 datasets that have been in used in this project
 65 | * Column 1: Statement (News headline or text).
 66 | * Column 2: Label (Label class contains: True, False)
 67 |  
 68 | You will see that newly created dataset has only 2 classes as compared to 6 from original classes. Below is method used for reducing the number of classes.
 69 | 
 70 | * Original 	--	New
 71 | * True	   	--	True
 72 | * Mostly-true	-- 	True
 73 | * Half-true	-- 	True
 74 | * Barely-true	-- 	False
 75 | * False		-- 	False
 76 | * Pants-fire	-- 	False
 77 | 
 78 | The dataset used for this project were in csv format named train.csv, test.csv and valid.csv and can be found in repo. The original datasets are in "liar" folder in tsv format.
 79 | 
 80 | 
 81 | ### File descriptions
 82 | 
 83 | #### DataPrep.py
 84 | This file contains all the pre processing functions needed to process all input documents and texts. First we read the train, test and validation data files then performed some pre processing like tokenizing, stemming etc. There are some exploratory data analysis is performed like response variable distribution and data quality checks like null or missing values etc.
 85 | 
 86 | #### FeatureSelection.py
 87 | In this file we have performed feature extraction and selection methods from sci-kit learn python libraries. For feature selection, we have used methods like simple bag-of-words and n-grams and then term frequency like tf-tdf weighting. we have also used word2vec and POS tagging to extract the features, though POS tagging and word2vec has not been used at this point in the project.
 88 | 
 89 | #### classifier.py
 90 | Here we have build all the classifiers for predicting the fake news detection. The extracted features are fed into different classifiers. We have used Naive-bayes, Logistic Regression, Linear SVM, Stochastic gradient descent and Random forest classifiers from sklearn. Each of the extracted features were used in all of the classifiers. Once fitting the model, we compared the f1 score and checked the confusion matrix. After fitting all the classifiers, 2 best performing models were selected as candidate models for fake news classification. We have performed parameter tuning by implementing GridSearchCV methods on these candidate models and chosen best performing parameters for these classifier. Finally selected model was used for fake news detection with the probability of truth. In Addition to this, We have also extracted the top 50 features from our term-frequency tfidf vectorizer to see what words are most and important in each of the classes. We have also used Precision-Recall and learning curves to see how training and test set performs when we increase the amount of data in our classifiers.
 91 | 
 92 | #### prediction.py
 93 | Our finally selected and best performing classifier was ```Logistic Regression``` which was then saved on disk with name ```final_model.sav```. Once you close this repository, this model will be copied to user's machine and will be used by prediction.py file to classify the fake news. It takes an news article as input from user then model is used for final classification output that is shown to user along with probability of truth.
 94 | 
 95 | Below is the Process Flow of the project:
 96 | 
 97 | <p align="center">
 98 |   <img width="600" height="750" src="https://github.com/nishitpatel01/Fake_News_Detection/blob/master/images/ProcessFlow.PNG">
 99 | </p>
100 | 
101 | ### Performance
102 | Below is the learning curves for our candidate models. 
103 | 
104 | **Logistic Regression Classifier**
105 | 
106 | <p align="center">
107 |   <img width="550" height="450" src="https://github.com/nishitpatel01/Fake_News_Detection/blob/master/images/LR_LCurve.PNG">
108 | </p>
109 | 
110 | **Random Forest Classifier**
111 | 
112 | <p align="center">
113 |   <img width="550" height="450" src="https://github.com/nishitpatel01/Fake_News_Detection/blob/master/images/RF_LCurve.png">
114 | </p>
115 | 
116 | ### Next steps
117 | As we can see that our best performing models had an f1 score in the range of 70's. This is due to less number of data that we have used for training purposes and simplicity of our models. For the future implementations, we could introduce some more feature selection methods such as POS tagging, word2vec and topic modeling. In addition, we could also increase the training data size. We will extend this project to implement these techniques in future to increase the accuracy and performance of our models.
118 | 
119 | 
120 | ### Installing and steps to run the software
121 | 
122 | A step by step series of examples that tell you have to get a development env running
123 | 
124 | 1. The first step would be to clone this repo in a folder in your local machine. To do that you need to run following command in command prompt or in git bash
125 | ```
126 | $ git clone https://github.com/nishitpatel01/Fake_News_Detection.git
127 | ```
128 | 
129 | 2. This will copy all the data source file, program files and model into your machine.
130 | 
131 | 3.
132 |    - If you have chosen to install anaconda then follow below instructions
133 |      - After all the files are saved in a folder in your machine. If you chosen to install anaconda from the steps given in 	               ```Prerequisites``` sections then open the anaconda prompt, change the directory to the folder where this project is saved in     your machine and type below command and press enter.
134 | 	```
135 | 	cd C:/your cloned project folder path goes here/
136 | 	```
137 |      - Once you are inside the directory call the ```prediction.py``` file, To do this, run below command in anaconda prompt.
138 | 	```
139 | 	python prediction.py
140 | 	```
141 |      - After hitting the enter, program will ask for an input which will be a piece of information or a news headline that you 	    	   want to verify. Once you paste or type news headline, then press enter.
142 | 
143 |      - Once you hit the enter, program will take user input (news headline) and will be used by model to classify in one of  categories of "True" and "False". Along with classifying the news headline, model will also provide a probability of truth associated with it.
144 | 
145 | 4.  If you have chosen to install python (and did not set up PATH variable for it) then follow below instructions:
146 |     - After you clone the project in a folder in your machine. Open command prompt and change the directory to project directory by running below command.
147 |     ```
148 |     cd C:/your cloned project folder path goes here/
149 |     ```
150 |     - Locate ```python.exe``` in your machine. you can search this in window explorer search bar. 
151 |     - Once you locate the ```python.exe``` path, you need to write whole path of it and then entire path of project folder with ```prediction.py``` at the end. For example if your ```python.exe``` is located at ```c:/Python36/python.exe``` and project folder is at ```c:/users/user_name/desktop/fake_news_detection/```, then your command to run program will be as below:
152 |     ```
153 |     c:/Python36/python.exe C:/users/user_name/desktop/fake_news_detection/prediction.py
154 |     ```
155 |     - After hitting the enter, program will ask for an input which will be a piece of information or a news headline that you 	    	   want to verify. Once you paste or type news headline, then press enter.
156 | 
157 |     - Once you hit the enter, program will take user input (news headline) and will be used by model to classify in one of  categories of "True" and "False". Along with classifying the news headline, model will also provide a probability of truth associated with it. It might take few seconds for model to classify the given statement so wait for it.
158 | 
159 | 5.  If you have chosen to install python (and already setup PATH variable for ```python.exe```) then follow instructions:
160 |     - Open the command prompt and change the directory to project folder as mentioned in above by running below command
161 |     ```
162 |     cd C:/your cloned project folder path goes here/
163 |     ```
164 |     - run below command
165 |     ```
166 |     python.exe C:/your cloned project folder path goes here/
167 |     ```
168 |     - After hitting the enter, program will ask for an input which will be a piece of information or a news headline that you 	    	   want to verify. Once you paste or type news headline, then press enter.
169 | 
170 |     - Once you hit the enter, program will take user input (news headline) and will be used by model to classify in one of  categories of "True" and "False". Along with classifying the news headline, model will also provide a probability of truth associated with it.
171 | 
172 | 


--------------------------------------------------------------------------------
/__pycache__/DataPrep.cpython-36.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nishitpatel01/Fake_News_Detection/880a2782ab627c10e4f9cf98c3d753d48439a9d1/__pycache__/DataPrep.cpython-36.pyc


--------------------------------------------------------------------------------
/__pycache__/FeatureSelection.cpython-36.pyc:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nishitpatel01/Fake_News_Detection/880a2782ab627c10e4f9cf98c3d753d48439a9d1/__pycache__/FeatureSelection.cpython-36.pyc


--------------------------------------------------------------------------------
/_config.yml:
--------------------------------------------------------------------------------
1 | theme: jekyll-theme-cayman


--------------------------------------------------------------------------------
/animate.css:
--------------------------------------------------------------------------------
 1 | @charset "UTF-8";
 2 | /*!
 3 | Animate.css - http://daneden.me/animate
 4 | Licensed under the MIT license
 5 | Copyright (c) 2013 Daniel Eden
 6 | Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
 7 | The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
 8 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
 9 | */
10 | 
11 | .animated{-webkit-animation-duration:1s;animation-duration:1s;-webkit-animation-fill-mode:both;animation-fill-mode:both}.animated.hinge{-webkit-animation-duration:2s;animation-duration:2s}@-webkit-keyframes bounce{0%,20%,50%,80%,100%{-webkit-transform:translateY(0);transform:translateY(0)}40%{-webkit-transform:translateY(-30px);transform:translateY(-30px)}60%{-webkit-transform:translateY(-15px);transform:translateY(-15px)}}@keyframes bounce{0%,20%,50%,80%,100%{-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}40%{-webkit-transform:translateY(-30px);-ms-transform:translateY(-30px);transform:translateY(-30px)}60%{-webkit-transform:translateY(-15px);-ms-transform:translateY(-15px);transform:translateY(-15px)}}.bounce{-webkit-animation-name:bounce;animation-name:bounce}@-webkit-keyframes flash{0%,50%,100%{opacity:1}25%,75%{opacity:0}}@keyframes flash{0%,50%,100%{opacity:1}25%,75%{opacity:0}}.flash{-webkit-animation-name:flash;animation-name:flash}@-webkit-keyframes pulse{0%{-webkit-transform:scale(1);transform:scale(1)}50%{-webkit-transform:scale(1.1);transform:scale(1.1)}100%{-webkit-transform:scale(1);transform:scale(1)}}@keyframes pulse{0%{-webkit-transform:scale(1);-ms-transform:scale(1);transform:scale(1)}50%{-webkit-transform:scale(1.1);-ms-transform:scale(1.1);transform:scale(1.1)}100%{-webkit-transform:scale(1);-ms-transform:scale(1);transform:scale(1)}}.pulse{-webkit-animation-name:pulse;animation-name:pulse}@-webkit-keyframes shake{0%,100%{-webkit-transform:translateX(0);transform:translateX(0)}10%,30%,50%,70%,90%{-webkit-transform:translateX(-10px);transform:translateX(-10px)}20%,40%,60%,80%{-webkit-transform:translateX(10px);transform:translateX(10px)}}@keyframes shake{0%,100%{-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}10%,30%,50%,70%,90%{-webkit-transform:translateX(-10px);-ms-transform:translateX(-10px);transform:translateX(-10px)}20%,40%,60%,80%{-webkit-transform:translateX(10px);-ms-transform:translateX(10px);transform:translateX(10px)}}.shake{-webkit-animation-name:shake;animation-name:shake}@-webkit-keyframes swing{20%{-webkit-transform:rotate(15deg);transform:rotate(15deg)}40%{-webkit-transform:rotate(-10deg);transform:rotate(-10deg)}60%{-webkit-transform:rotate(5deg);transform:rotate(5deg)}80%{-webkit-transform:rotate(-5deg);transform:rotate(-5deg)}100%{-webkit-transform:rotate(0deg);transform:rotate(0deg)}}@keyframes swing{20%{-webkit-transform:rotate(15deg);-ms-transform:rotate(15deg);transform:rotate(15deg)}40%{-webkit-transform:rotate(-10deg);-ms-transform:rotate(-10deg);transform:rotate(-10deg)}60%{-webkit-transform:rotate(5deg);-ms-transform:rotate(5deg);transform:rotate(5deg)}80%{-webkit-transform:rotate(-5deg);-ms-transform:rotate(-5deg);transform:rotate(-5deg)}100%{-webkit-transform:rotate(0deg);-ms-transform:rotate(0deg);transform:rotate(0deg)}}.swing{-webkit-transform-origin:top center;-ms-transform-origin:top center;transform-origin:top center;-webkit-animation-name:swing;animation-name:swing}@-webkit-keyframes tada{0%{-webkit-transform:scale(1);transform:scale(1)}10%,20%{-webkit-transform:scale(0.9) rotate(-3deg);transform:scale(0.9) rotate(-3deg)}30%,50%,70%,90%{-webkit-transform:scale(1.1) rotate(3deg);transform:scale(1.1) rotate(3deg)}40%,60%,80%{-webkit-transform:scale(1.1) rotate(-3deg);transform:scale(1.1) rotate(-3deg)}100%{-webkit-transform:scale(1) rotate(0);transform:scale(1) rotate(0)}}@keyframes tada{0%{-webkit-transform:scale(1);-ms-transform:scale(1);transform:scale(1)}10%,20%{-webkit-transform:scale(0.9) rotate(-3deg);-ms-transform:scale(0.9) rotate(-3deg);transform:scale(0.9) rotate(-3deg)}30%,50%,70%,90%{-webkit-transform:scale(1.1) rotate(3deg);-ms-transform:scale(1.1) rotate(3deg);transform:scale(1.1) rotate(3deg)}40%,60%,80%{-webkit-transform:scale(1.1) rotate(-3deg);-ms-transform:scale(1.1) rotate(-3deg);transform:scale(1.1) rotate(-3deg)}100%{-webkit-transform:scale(1) rotate(0);-ms-transform:scale(1) rotate(0);transform:scale(1) rotate(0)}}.tada{-webkit-animation-name:tada;animation-name:tada}@-webkit-keyframes wobble{0%{-webkit-transform:translateX(0%);transform:translateX(0%)}15%{-webkit-transform:translateX(-25%) rotate(-5deg);transform:translateX(-25%) rotate(-5deg)}30%{-webkit-transform:translateX(20%) rotate(3deg);transform:translateX(20%) rotate(3deg)}45%{-webkit-transform:translateX(-15%) rotate(-3deg);transform:translateX(-15%) rotate(-3deg)}60%{-webkit-transform:translateX(10%) rotate(2deg);transform:translateX(10%) rotate(2deg)}75%{-webkit-transform:translateX(-5%) rotate(-1deg);transform:translateX(-5%) rotate(-1deg)}100%{-webkit-transform:translateX(0%);transform:translateX(0%)}}@keyframes wobble{0%{-webkit-transform:translateX(0%);-ms-transform:translateX(0%);transform:translateX(0%)}15%{-webkit-transform:translateX(-25%) rotate(-5deg);-ms-transform:translateX(-25%) rotate(-5deg);transform:translateX(-25%) rotate(-5deg)}30%{-webkit-transform:translateX(20%) rotate(3deg);-ms-transform:translateX(20%) rotate(3deg);transform:translateX(20%) rotate(3deg)}45%{-webkit-transform:translateX(-15%) rotate(-3deg);-ms-transform:translateX(-15%) rotate(-3deg);transform:translateX(-15%) rotate(-3deg)}60%{-webkit-transform:translateX(10%) rotate(2deg);-ms-transform:translateX(10%) rotate(2deg);transform:translateX(10%) rotate(2deg)}75%{-webkit-transform:translateX(-5%) rotate(-1deg);-ms-transform:translateX(-5%) rotate(-1deg);transform:translateX(-5%) rotate(-1deg)}100%{-webkit-transform:translateX(0%);-ms-transform:translateX(0%);transform:translateX(0%)}}.wobble{-webkit-animation-name:wobble;animation-name:wobble}@-webkit-keyframes bounceIn{0%{opacity:0;-webkit-transform:scale(.3);transform:scale(.3)}50%{opacity:1;-webkit-transform:scale(1.05);transform:scale(1.05)}70%{-webkit-transform:scale(.9);transform:scale(.9)}100%{-webkit-transform:scale(1);transform:scale(1)}}@keyframes bounceIn{0%{opacity:0;-webkit-transform:scale(.3);-ms-transform:scale(.3);transform:scale(.3)}50%{opacity:1;-webkit-transform:scale(1.05);-ms-transform:scale(1.05);transform:scale(1.05)}70%{-webkit-transform:scale(.9);-ms-transform:scale(.9);transform:scale(.9)}100%{-webkit-transform:scale(1);-ms-transform:scale(1);transform:scale(1)}}.bounceIn{-webkit-animation-name:bounceIn;animation-name:bounceIn}@-webkit-keyframes bounceInDown{0%{opacity:0;-webkit-transform:translateY(-2000px);transform:translateY(-2000px)}60%{opacity:1;-webkit-transform:translateY(30px);transform:translateY(30px)}80%{-webkit-transform:translateY(-10px);transform:translateY(-10px)}100%{-webkit-transform:translateY(0);transform:translateY(0)}}@keyframes bounceInDown{0%{opacity:0;-webkit-transform:translateY(-2000px);-ms-transform:translateY(-2000px);transform:translateY(-2000px)}60%{opacity:1;-webkit-transform:translateY(30px);-ms-transform:translateY(30px);transform:translateY(30px)}80%{-webkit-transform:translateY(-10px);-ms-transform:translateY(-10px);transform:translateY(-10px)}100%{-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}}.bounceInDown{-webkit-animation-name:bounceInDown;animation-name:bounceInDown}@-webkit-keyframes bounceInLeft{0%{opacity:0;-webkit-transform:translateX(-2000px);transform:translateX(-2000px)}60%{opacity:1;-webkit-transform:translateX(30px);transform:translateX(30px)}80%{-webkit-transform:translateX(-10px);transform:translateX(-10px)}100%{-webkit-transform:translateX(0);transform:translateX(0)}}@keyframes bounceInLeft{0%{opacity:0;-webkit-transform:translateX(-2000px);-ms-transform:translateX(-2000px);transform:translateX(-2000px)}60%{opacity:1;-webkit-transform:translateX(30px);-ms-transform:translateX(30px);transform:translateX(30px)}80%{-webkit-transform:translateX(-10px);-ms-transform:translateX(-10px);transform:translateX(-10px)}100%{-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}}.bounceInLeft{-webkit-animation-name:bounceInLeft;animation-name:bounceInLeft}@-webkit-keyframes bounceInRight{0%{opacity:0;-webkit-transform:translateX(2000px);transform:translateX(2000px)}60%{opacity:1;-webkit-transform:translateX(-30px);transform:translateX(-30px)}80%{-webkit-transform:translateX(10px);transform:translateX(10px)}100%{-webkit-transform:translateX(0);transform:translateX(0)}}@keyframes bounceInRight{0%{opacity:0;-webkit-transform:translateX(2000px);-ms-transform:translateX(2000px);transform:translateX(2000px)}60%{opacity:1;-webkit-transform:translateX(-30px);-ms-transform:translateX(-30px);transform:translateX(-30px)}80%{-webkit-transform:translateX(10px);-ms-transform:translateX(10px);transform:translateX(10px)}100%{-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}}.bounceInRight{-webkit-animation-name:bounceInRight;animation-name:bounceInRight}@-webkit-keyframes bounceInUp{0%{opacity:0;-webkit-transform:translateY(2000px);transform:translateY(2000px)}60%{opacity:1;-webkit-transform:translateY(-30px);transform:translateY(-30px)}80%{-webkit-transform:translateY(10px);transform:translateY(10px)}100%{-webkit-transform:translateY(0);transform:translateY(0)}}@keyframes bounceInUp{0%{opacity:0;-webkit-transform:translateY(2000px);-ms-transform:translateY(2000px);transform:translateY(2000px)}60%{opacity:1;-webkit-transform:translateY(-30px);-ms-transform:translateY(-30px);transform:translateY(-30px)}80%{-webkit-transform:translateY(10px);-ms-transform:translateY(10px);transform:translateY(10px)}100%{-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}}.bounceInUp{-webkit-animation-name:bounceInUp;animation-name:bounceInUp}@-webkit-keyframes bounceOut{0%{-webkit-transform:scale(1);transform:scale(1)}25%{-webkit-transform:scale(.95);transform:scale(.95)}50%{opacity:1;-webkit-transform:scale(1.1);transform:scale(1.1)}100%{opacity:0;-webkit-transform:scale(.3);transform:scale(.3)}}@keyframes bounceOut{0%{-webkit-transform:scale(1);-ms-transform:scale(1);transform:scale(1)}25%{-webkit-transform:scale(.95);-ms-transform:scale(.95);transform:scale(.95)}50%{opacity:1;-webkit-transform:scale(1.1);-ms-transform:scale(1.1);transform:scale(1.1)}100%{opacity:0;-webkit-transform:scale(.3);-ms-transform:scale(.3);transform:scale(.3)}}.bounceOut{-webkit-animation-name:bounceOut;animation-name:bounceOut}@-webkit-keyframes bounceOutDown{0%{-webkit-transform:translateY(0);transform:translateY(0)}20%{opacity:1;-webkit-transform:translateY(-20px);transform:translateY(-20px)}100%{opacity:0;-webkit-transform:translateY(2000px);transform:translateY(2000px)}}@keyframes bounceOutDown{0%{-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}20%{opacity:1;-webkit-transform:translateY(-20px);-ms-transform:translateY(-20px);transform:translateY(-20px)}100%{opacity:0;-webkit-transform:translateY(2000px);-ms-transform:translateY(2000px);transform:translateY(2000px)}}.bounceOutDown{-webkit-animation-name:bounceOutDown;animation-name:bounceOutDown}@-webkit-keyframes bounceOutLeft{0%{-webkit-transform:translateX(0);transform:translateX(0)}20%{opacity:1;-webkit-transform:translateX(20px);transform:translateX(20px)}100%{opacity:0;-webkit-transform:translateX(-2000px);transform:translateX(-2000px)}}@keyframes bounceOutLeft{0%{-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}20%{opacity:1;-webkit-transform:translateX(20px);-ms-transform:translateX(20px);transform:translateX(20px)}100%{opacity:0;-webkit-transform:translateX(-2000px);-ms-transform:translateX(-2000px);transform:translateX(-2000px)}}.bounceOutLeft{-webkit-animation-name:bounceOutLeft;animation-name:bounceOutLeft}@-webkit-keyframes bounceOutRight{0%{-webkit-transform:translateX(0);transform:translateX(0)}20%{opacity:1;-webkit-transform:translateX(-20px);transform:translateX(-20px)}100%{opacity:0;-webkit-transform:translateX(2000px);transform:translateX(2000px)}}@keyframes bounceOutRight{0%{-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}20%{opacity:1;-webkit-transform:translateX(-20px);-ms-transform:translateX(-20px);transform:translateX(-20px)}100%{opacity:0;-webkit-transform:translateX(2000px);-ms-transform:translateX(2000px);transform:translateX(2000px)}}.bounceOutRight{-webkit-animation-name:bounceOutRight;animation-name:bounceOutRight}@-webkit-keyframes bounceOutUp{0%{-webkit-transform:translateY(0);transform:translateY(0)}20%{opacity:1;-webkit-transform:translateY(20px);transform:translateY(20px)}100%{opacity:0;-webkit-transform:translateY(-2000px);transform:translateY(-2000px)}}@keyframes bounceOutUp{0%{-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}20%{opacity:1;-webkit-transform:translateY(20px);-ms-transform:translateY(20px);transform:translateY(20px)}100%{opacity:0;-webkit-transform:translateY(-2000px);-ms-transform:translateY(-2000px);transform:translateY(-2000px)}}.bounceOutUp{-webkit-animation-name:bounceOutUp;animation-name:bounceOutUp}@-webkit-keyframes fadeIn{0%{opacity:0}100%{opacity:1}}@keyframes fadeIn{0%{opacity:0}100%{opacity:1}}.fadeIn{-webkit-animation-name:fadeIn;animation-name:fadeIn}@-webkit-keyframes fadeInDown{0%{opacity:0;-webkit-transform:translateY(-20px);transform:translateY(-20px)}100%{opacity:1;-webkit-transform:translateY(0);transform:translateY(0)}}@keyframes fadeInDown{0%{opacity:0;-webkit-transform:translateY(-20px);-ms-transform:translateY(-20px);transform:translateY(-20px)}100%{opacity:1;-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}}.fadeInDown{-webkit-animation-name:fadeInDown;animation-name:fadeInDown}@-webkit-keyframes fadeInDownBig{0%{opacity:0;-webkit-transform:translateY(-2000px);transform:translateY(-2000px)}100%{opacity:1;-webkit-transform:translateY(0);transform:translateY(0)}}@keyframes fadeInDownBig{0%{opacity:0;-webkit-transform:translateY(-2000px);-ms-transform:translateY(-2000px);transform:translateY(-2000px)}100%{opacity:1;-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}}.fadeInDownBig{-webkit-animation-name:fadeInDownBig;animation-name:fadeInDownBig}@-webkit-keyframes fadeInLeft{0%{opacity:0;-webkit-transform:translateX(-20px);transform:translateX(-20px)}100%{opacity:1;-webkit-transform:translateX(0);transform:translateX(0)}}@keyframes fadeInLeft{0%{opacity:0;-webkit-transform:translateX(-20px);-ms-transform:translateX(-20px);transform:translateX(-20px)}100%{opacity:1;-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}}.fadeInLeft{-webkit-animation-name:fadeInLeft;animation-name:fadeInLeft}@-webkit-keyframes fadeInLeftBig{0%{opacity:0;-webkit-transform:translateX(-2000px);transform:translateX(-2000px)}100%{opacity:1;-webkit-transform:translateX(0);transform:translateX(0)}}@keyframes fadeInLeftBig{0%{opacity:0;-webkit-transform:translateX(-2000px);-ms-transform:translateX(-2000px);transform:translateX(-2000px)}100%{opacity:1;-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}}.fadeInLeftBig{-webkit-animation-name:fadeInLeftBig;animation-name:fadeInLeftBig}@-webkit-keyframes fadeInRight{0%{opacity:0;-webkit-transform:translateX(20px);transform:translateX(20px)}100%{opacity:1;-webkit-transform:translateX(0);transform:translateX(0)}}@keyframes fadeInRight{0%{opacity:0;-webkit-transform:translateX(20px);-ms-transform:translateX(20px);transform:translateX(20px)}100%{opacity:1;-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}}.fadeInRight{-webkit-animation-name:fadeInRight;animation-name:fadeInRight}@-webkit-keyframes fadeInRightBig{0%{opacity:0;-webkit-transform:translateX(2000px);transform:translateX(2000px)}100%{opacity:1;-webkit-transform:translateX(0);transform:translateX(0)}}@keyframes fadeInRightBig{0%{opacity:0;-webkit-transform:translateX(2000px);-ms-transform:translateX(2000px);transform:translateX(2000px)}100%{opacity:1;-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}}.fadeInRightBig{-webkit-animation-name:fadeInRightBig;animation-name:fadeInRightBig}@-webkit-keyframes fadeInUp{0%{opacity:0;-webkit-transform:translateY(20px);transform:translateY(20px)}100%{opacity:1;-webkit-transform:translateY(0);transform:translateY(0)}}@keyframes fadeInUp{0%{opacity:0;-webkit-transform:translateY(20px);-ms-transform:translateY(20px);transform:translateY(20px)}100%{opacity:1;-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}}.fadeInUp{-webkit-animation-name:fadeInUp;animation-name:fadeInUp}@-webkit-keyframes fadeInUpBig{0%{opacity:0;-webkit-transform:translateY(2000px);transform:translateY(2000px)}100%{opacity:1;-webkit-transform:translateY(0);transform:translateY(0)}}@keyframes fadeInUpBig{0%{opacity:0;-webkit-transform:translateY(2000px);-ms-transform:translateY(2000px);transform:translateY(2000px)}100%{opacity:1;-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}}.fadeInUpBig{-webkit-animation-name:fadeInUpBig;animation-name:fadeInUpBig}@-webkit-keyframes fadeOut{0%{opacity:1}100%{opacity:0}}@keyframes fadeOut{0%{opacity:1}100%{opacity:0}}.fadeOut{-webkit-animation-name:fadeOut;animation-name:fadeOut}@-webkit-keyframes fadeOutDown{0%{opacity:1;-webkit-transform:translateY(0);transform:translateY(0)}100%{opacity:0;-webkit-transform:translateY(20px);transform:translateY(20px)}}@keyframes fadeOutDown{0%{opacity:1;-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}100%{opacity:0;-webkit-transform:translateY(20px);-ms-transform:translateY(20px);transform:translateY(20px)}}.fadeOutDown{-webkit-animation-name:fadeOutDown;animation-name:fadeOutDown}@-webkit-keyframes fadeOutDownBig{0%{opacity:1;-webkit-transform:translateY(0);transform:translateY(0)}100%{opacity:0;-webkit-transform:translateY(2000px);transform:translateY(2000px)}}@keyframes fadeOutDownBig{0%{opacity:1;-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}100%{opacity:0;-webkit-transform:translateY(2000px);-ms-transform:translateY(2000px);transform:translateY(2000px)}}.fadeOutDownBig{-webkit-animation-name:fadeOutDownBig;animation-name:fadeOutDownBig}@-webkit-keyframes fadeOutLeft{0%{opacity:1;-webkit-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(-20px);transform:translateX(-20px)}}@keyframes fadeOutLeft{0%{opacity:1;-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(-20px);-ms-transform:translateX(-20px);transform:translateX(-20px)}}.fadeOutLeft{-webkit-animation-name:fadeOutLeft;animation-name:fadeOutLeft}@-webkit-keyframes fadeOutLeftBig{0%{opacity:1;-webkit-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(-2000px);transform:translateX(-2000px)}}@keyframes fadeOutLeftBig{0%{opacity:1;-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(-2000px);-ms-transform:translateX(-2000px);transform:translateX(-2000px)}}.fadeOutLeftBig{-webkit-animation-name:fadeOutLeftBig;animation-name:fadeOutLeftBig}@-webkit-keyframes fadeOutRight{0%{opacity:1;-webkit-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(20px);transform:translateX(20px)}}@keyframes fadeOutRight{0%{opacity:1;-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(20px);-ms-transform:translateX(20px);transform:translateX(20px)}}.fadeOutRight{-webkit-animation-name:fadeOutRight;animation-name:fadeOutRight}@-webkit-keyframes fadeOutRightBig{0%{opacity:1;-webkit-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(2000px);transform:translateX(2000px)}}@keyframes fadeOutRightBig{0%{opacity:1;-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(2000px);-ms-transform:translateX(2000px);transform:translateX(2000px)}}.fadeOutRightBig{-webkit-animation-name:fadeOutRightBig;animation-name:fadeOutRightBig}@-webkit-keyframes fadeOutUp{0%{opacity:1;-webkit-transform:translateY(0);transform:translateY(0)}100%{opacity:0;-webkit-transform:translateY(-20px);transform:translateY(-20px)}}@keyframes fadeOutUp{0%{opacity:1;-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}100%{opacity:0;-webkit-transform:translateY(-20px);-ms-transform:translateY(-20px);transform:translateY(-20px)}}.fadeOutUp{-webkit-animation-name:fadeOutUp;animation-name:fadeOutUp}@-webkit-keyframes fadeOutUpBig{0%{opacity:1;-webkit-transform:translateY(0);transform:translateY(0)}100%{opacity:0;-webkit-transform:translateY(-2000px);transform:translateY(-2000px)}}@keyframes fadeOutUpBig{0%{opacity:1;-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}100%{opacity:0;-webkit-transform:translateY(-2000px);-ms-transform:translateY(-2000px);transform:translateY(-2000px)}}.fadeOutUpBig{-webkit-animation-name:fadeOutUpBig;animation-name:fadeOutUpBig}@-webkit-keyframes flip{0%{-webkit-transform:perspective(400px) translateZ(0) rotateY(0) scale(1);transform:perspective(400px) translateZ(0) rotateY(0) scale(1);-webkit-animation-timing-function:ease-out;animation-timing-function:ease-out}40%{-webkit-transform:perspective(400px) translateZ(150px) rotateY(170deg) scale(1);transform:perspective(400px) translateZ(150px) rotateY(170deg) scale(1);-webkit-animation-timing-function:ease-out;animation-timing-function:ease-out}50%{-webkit-transform:perspective(400px) translateZ(150px) rotateY(190deg) scale(1);transform:perspective(400px) translateZ(150px) rotateY(190deg) scale(1);-webkit-animation-timing-function:ease-in;animation-timing-function:ease-in}80%{-webkit-transform:perspective(400px) translateZ(0) rotateY(360deg) scale(.95);transform:perspective(400px) translateZ(0) rotateY(360deg) scale(.95);-webkit-animation-timing-function:ease-in;animation-timing-function:ease-in}100%{-webkit-transform:perspective(400px) translateZ(0) rotateY(360deg) scale(1);transform:perspective(400px) translateZ(0) rotateY(360deg) scale(1);-webkit-animation-timing-function:ease-in;animation-timing-function:ease-in}}@keyframes flip{0%{-webkit-transform:perspective(400px) translateZ(0) rotateY(0) scale(1);-ms-transform:perspective(400px) translateZ(0) rotateY(0) scale(1);transform:perspective(400px) translateZ(0) rotateY(0) scale(1);-webkit-animation-timing-function:ease-out;animation-timing-function:ease-out}40%{-webkit-transform:perspective(400px) translateZ(150px) rotateY(170deg) scale(1);-ms-transform:perspective(400px) translateZ(150px) rotateY(170deg) scale(1);transform:perspective(400px) translateZ(150px) rotateY(170deg) scale(1);-webkit-animation-timing-function:ease-out;animation-timing-function:ease-out}50%{-webkit-transform:perspective(400px) translateZ(150px) rotateY(190deg) scale(1);-ms-transform:perspective(400px) translateZ(150px) rotateY(190deg) scale(1);transform:perspective(400px) translateZ(150px) rotateY(190deg) scale(1);-webkit-animation-timing-function:ease-in;animation-timing-function:ease-in}80%{-webkit-transform:perspective(400px) translateZ(0) rotateY(360deg) scale(.95);-ms-transform:perspective(400px) translateZ(0) rotateY(360deg) scale(.95);transform:perspective(400px) translateZ(0) rotateY(360deg) scale(.95);-webkit-animation-timing-function:ease-in;animation-timing-function:ease-in}100%{-webkit-transform:perspective(400px) translateZ(0) rotateY(360deg) scale(1);-ms-transform:perspective(400px) translateZ(0) rotateY(360deg) scale(1);transform:perspective(400px) translateZ(0) rotateY(360deg) scale(1);-webkit-animation-timing-function:ease-in;animation-timing-function:ease-in}}.animated.flip{-webkit-backface-visibility:visible;-ms-backface-visibility:visible;backface-visibility:visible;-webkit-animation-name:flip;animation-name:flip}@-webkit-keyframes flipInX{0%{-webkit-transform:perspective(400px) rotateX(90deg);transform:perspective(400px) rotateX(90deg);opacity:0}40%{-webkit-transform:perspective(400px) rotateX(-10deg);transform:perspective(400px) rotateX(-10deg)}70%{-webkit-transform:perspective(400px) rotateX(10deg);transform:perspective(400px) rotateX(10deg)}100%{-webkit-transform:perspective(400px) rotateX(0deg);transform:perspective(400px) rotateX(0deg);opacity:1}}@keyframes flipInX{0%{-webkit-transform:perspective(400px) rotateX(90deg);-ms-transform:perspective(400px) rotateX(90deg);transform:perspective(400px) rotateX(90deg);opacity:0}40%{-webkit-transform:perspective(400px) rotateX(-10deg);-ms-transform:perspective(400px) rotateX(-10deg);transform:perspective(400px) rotateX(-10deg)}70%{-webkit-transform:perspective(400px) rotateX(10deg);-ms-transform:perspective(400px) rotateX(10deg);transform:perspective(400px) rotateX(10deg)}100%{-webkit-transform:perspective(400px) rotateX(0deg);-ms-transform:perspective(400px) rotateX(0deg);transform:perspective(400px) rotateX(0deg);opacity:1}}.flipInX{-webkit-backface-visibility:visible !important;-ms-backface-visibility:visible !important;backface-visibility:visible !important;-webkit-animation-name:flipInX;animation-name:flipInX}@-webkit-keyframes flipInY{0%{-webkit-transform:perspective(400px) rotateY(90deg);transform:perspective(400px) rotateY(90deg);opacity:0}40%{-webkit-transform:perspective(400px) rotateY(-10deg);transform:perspective(400px) rotateY(-10deg)}70%{-webkit-transform:perspective(400px) rotateY(10deg);transform:perspective(400px) rotateY(10deg)}100%{-webkit-transform:perspective(400px) rotateY(0deg);transform:perspective(400px) rotateY(0deg);opacity:1}}@keyframes flipInY{0%{-webkit-transform:perspective(400px) rotateY(90deg);-ms-transform:perspective(400px) rotateY(90deg);transform:perspective(400px) rotateY(90deg);opacity:0}40%{-webkit-transform:perspective(400px) rotateY(-10deg);-ms-transform:perspective(400px) rotateY(-10deg);transform:perspective(400px) rotateY(-10deg)}70%{-webkit-transform:perspective(400px) rotateY(10deg);-ms-transform:perspective(400px) rotateY(10deg);transform:perspective(400px) rotateY(10deg)}100%{-webkit-transform:perspective(400px) rotateY(0deg);-ms-transform:perspective(400px) rotateY(0deg);transform:perspective(400px) rotateY(0deg);opacity:1}}.flipInY{-webkit-backface-visibility:visible !important;-ms-backface-visibility:visible !important;backface-visibility:visible !important;-webkit-animation-name:flipInY;animation-name:flipInY}@-webkit-keyframes flipOutX{0%{-webkit-transform:perspective(400px) rotateX(0deg);transform:perspective(400px) rotateX(0deg);opacity:1}100%{-webkit-transform:perspective(400px) rotateX(90deg);transform:perspective(400px) rotateX(90deg);opacity:0}}@keyframes flipOutX{0%{-webkit-transform:perspective(400px) rotateX(0deg);-ms-transform:perspective(400px) rotateX(0deg);transform:perspective(400px) rotateX(0deg);opacity:1}100%{-webkit-transform:perspective(400px) rotateX(90deg);-ms-transform:perspective(400px) rotateX(90deg);transform:perspective(400px) rotateX(90deg);opacity:0}}.flipOutX{-webkit-animation-name:flipOutX;animation-name:flipOutX;-webkit-backface-visibility:visible !important;-ms-backface-visibility:visible !important;backface-visibility:visible !important}@-webkit-keyframes flipOutY{0%{-webkit-transform:perspective(400px) rotateY(0deg);transform:perspective(400px) rotateY(0deg);opacity:1}100%{-webkit-transform:perspective(400px) rotateY(90deg);transform:perspective(400px) rotateY(90deg);opacity:0}}@keyframes flipOutY{0%{-webkit-transform:perspective(400px) rotateY(0deg);-ms-transform:perspective(400px) rotateY(0deg);transform:perspective(400px) rotateY(0deg);opacity:1}100%{-webkit-transform:perspective(400px) rotateY(90deg);-ms-transform:perspective(400px) rotateY(90deg);transform:perspective(400px) rotateY(90deg);opacity:0}}.flipOutY{-webkit-backface-visibility:visible !important;-ms-backface-visibility:visible !important;backface-visibility:visible !important;-webkit-animation-name:flipOutY;animation-name:flipOutY}@-webkit-keyframes lightSpeedIn{0%{-webkit-transform:translateX(100%) skewX(-30deg);transform:translateX(100%) skewX(-30deg);opacity:0}60%{-webkit-transform:translateX(-20%) skewX(30deg);transform:translateX(-20%) skewX(30deg);opacity:1}80%{-webkit-transform:translateX(0%) skewX(-15deg);transform:translateX(0%) skewX(-15deg);opacity:1}100%{-webkit-transform:translateX(0%) skewX(0deg);transform:translateX(0%) skewX(0deg);opacity:1}}@keyframes lightSpeedIn{0%{-webkit-transform:translateX(100%) skewX(-30deg);-ms-transform:translateX(100%) skewX(-30deg);transform:translateX(100%) skewX(-30deg);opacity:0}60%{-webkit-transform:translateX(-20%) skewX(30deg);-ms-transform:translateX(-20%) skewX(30deg);transform:translateX(-20%) skewX(30deg);opacity:1}80%{-webkit-transform:translateX(0%) skewX(-15deg);-ms-transform:translateX(0%) skewX(-15deg);transform:translateX(0%) skewX(-15deg);opacity:1}100%{-webkit-transform:translateX(0%) skewX(0deg);-ms-transform:translateX(0%) skewX(0deg);transform:translateX(0%) skewX(0deg);opacity:1}}.lightSpeedIn{-webkit-animation-name:lightSpeedIn;animation-name:lightSpeedIn;-webkit-animation-timing-function:ease-out;animation-timing-function:ease-out}@-webkit-keyframes lightSpeedOut{0%{-webkit-transform:translateX(0%) skewX(0deg);transform:translateX(0%) skewX(0deg);opacity:1}100%{-webkit-transform:translateX(100%) skewX(-30deg);transform:translateX(100%) skewX(-30deg);opacity:0}}@keyframes lightSpeedOut{0%{-webkit-transform:translateX(0%) skewX(0deg);-ms-transform:translateX(0%) skewX(0deg);transform:translateX(0%) skewX(0deg);opacity:1}100%{-webkit-transform:translateX(100%) skewX(-30deg);-ms-transform:translateX(100%) skewX(-30deg);transform:translateX(100%) skewX(-30deg);opacity:0}}.lightSpeedOut{-webkit-animation-name:lightSpeedOut;animation-name:lightSpeedOut;-webkit-animation-timing-function:ease-in;animation-timing-function:ease-in}@-webkit-keyframes rotateIn{0%{-webkit-transform-origin:center center;transform-origin:center center;-webkit-transform:rotate(-200deg);transform:rotate(-200deg);opacity:0}100%{-webkit-transform-origin:center center;transform-origin:center center;-webkit-transform:rotate(0);transform:rotate(0);opacity:1}}@keyframes rotateIn{0%{-webkit-transform-origin:center center;-ms-transform-origin:center center;transform-origin:center center;-webkit-transform:rotate(-200deg);-ms-transform:rotate(-200deg);transform:rotate(-200deg);opacity:0}100%{-webkit-transform-origin:center center;-ms-transform-origin:center center;transform-origin:center center;-webkit-transform:rotate(0);-ms-transform:rotate(0);transform:rotate(0);opacity:1}}.rotateIn{-webkit-animation-name:rotateIn;animation-name:rotateIn}@-webkit-keyframes rotateInDownLeft{0%{-webkit-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(-90deg);transform:rotate(-90deg);opacity:0}100%{-webkit-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(0);transform:rotate(0);opacity:1}}@keyframes rotateInDownLeft{0%{-webkit-transform-origin:left bottom;-ms-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(-90deg);-ms-transform:rotate(-90deg);transform:rotate(-90deg);opacity:0}100%{-webkit-transform-origin:left bottom;-ms-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(0);-ms-transform:rotate(0);transform:rotate(0);opacity:1}}.rotateInDownLeft{-webkit-animation-name:rotateInDownLeft;animation-name:rotateInDownLeft}@-webkit-keyframes rotateInDownRight{0%{-webkit-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(90deg);transform:rotate(90deg);opacity:0}100%{-webkit-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(0);transform:rotate(0);opacity:1}}@keyframes rotateInDownRight{0%{-webkit-transform-origin:right bottom;-ms-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(90deg);-ms-transform:rotate(90deg);transform:rotate(90deg);opacity:0}100%{-webkit-transform-origin:right bottom;-ms-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(0);-ms-transform:rotate(0);transform:rotate(0);opacity:1}}.rotateInDownRight{-webkit-animation-name:rotateInDownRight;animation-name:rotateInDownRight}@-webkit-keyframes rotateInUpLeft{0%{-webkit-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(90deg);transform:rotate(90deg);opacity:0}100%{-webkit-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(0);transform:rotate(0);opacity:1}}@keyframes rotateInUpLeft{0%{-webkit-transform-origin:left bottom;-ms-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(90deg);-ms-transform:rotate(90deg);transform:rotate(90deg);opacity:0}100%{-webkit-transform-origin:left bottom;-ms-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(0);-ms-transform:rotate(0);transform:rotate(0);opacity:1}}.rotateInUpLeft{-webkit-animation-name:rotateInUpLeft;animation-name:rotateInUpLeft}@-webkit-keyframes rotateInUpRight{0%{-webkit-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(-90deg);transform:rotate(-90deg);opacity:0}100%{-webkit-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(0);transform:rotate(0);opacity:1}}@keyframes rotateInUpRight{0%{-webkit-transform-origin:right bottom;-ms-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(-90deg);-ms-transform:rotate(-90deg);transform:rotate(-90deg);opacity:0}100%{-webkit-transform-origin:right bottom;-ms-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(0);-ms-transform:rotate(0);transform:rotate(0);opacity:1}}.rotateInUpRight{-webkit-animation-name:rotateInUpRight;animation-name:rotateInUpRight}@-webkit-keyframes rotateOut{0%{-webkit-transform-origin:center center;transform-origin:center center;-webkit-transform:rotate(0);transform:rotate(0);opacity:1}100%{-webkit-transform-origin:center center;transform-origin:center center;-webkit-transform:rotate(200deg);transform:rotate(200deg);opacity:0}}@keyframes rotateOut{0%{-webkit-transform-origin:center center;-ms-transform-origin:center center;transform-origin:center center;-webkit-transform:rotate(0);-ms-transform:rotate(0);transform:rotate(0);opacity:1}100%{-webkit-transform-origin:center center;-ms-transform-origin:center center;transform-origin:center center;-webkit-transform:rotate(200deg);-ms-transform:rotate(200deg);transform:rotate(200deg);opacity:0}}.rotateOut{-webkit-animation-name:rotateOut;animation-name:rotateOut}@-webkit-keyframes rotateOutDownLeft{0%{-webkit-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(0);transform:rotate(0);opacity:1}100%{-webkit-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(90deg);transform:rotate(90deg);opacity:0}}@keyframes rotateOutDownLeft{0%{-webkit-transform-origin:left bottom;-ms-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(0);-ms-transform:rotate(0);transform:rotate(0);opacity:1}100%{-webkit-transform-origin:left bottom;-ms-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(90deg);-ms-transform:rotate(90deg);transform:rotate(90deg);opacity:0}}.rotateOutDownLeft{-webkit-animation-name:rotateOutDownLeft;animation-name:rotateOutDownLeft}@-webkit-keyframes rotateOutDownRight{0%{-webkit-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(0);transform:rotate(0);opacity:1}100%{-webkit-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(-90deg);transform:rotate(-90deg);opacity:0}}@keyframes rotateOutDownRight{0%{-webkit-transform-origin:right bottom;-ms-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(0);-ms-transform:rotate(0);transform:rotate(0);opacity:1}100%{-webkit-transform-origin:right bottom;-ms-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(-90deg);-ms-transform:rotate(-90deg);transform:rotate(-90deg);opacity:0}}.rotateOutDownRight{-webkit-animation-name:rotateOutDownRight;animation-name:rotateOutDownRight}@-webkit-keyframes rotateOutUpLeft{0%{-webkit-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(0);transform:rotate(0);opacity:1}100%{-webkit-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(-90deg);transform:rotate(-90deg);opacity:0}}@keyframes rotateOutUpLeft{0%{-webkit-transform-origin:left bottom;-ms-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(0);-ms-transform:rotate(0);transform:rotate(0);opacity:1}100%{-webkit-transform-origin:left bottom;-ms-transform-origin:left bottom;transform-origin:left bottom;-webkit-transform:rotate(-90deg);-ms-transform:rotate(-90deg);transform:rotate(-90deg);opacity:0}}.rotateOutUpLeft{-webkit-animation-name:rotateOutUpLeft;animation-name:rotateOutUpLeft}@-webkit-keyframes rotateOutUpRight{0%{-webkit-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(0);transform:rotate(0);opacity:1}100%{-webkit-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(90deg);transform:rotate(90deg);opacity:0}}@keyframes rotateOutUpRight{0%{-webkit-transform-origin:right bottom;-ms-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(0);-ms-transform:rotate(0);transform:rotate(0);opacity:1}100%{-webkit-transform-origin:right bottom;-ms-transform-origin:right bottom;transform-origin:right bottom;-webkit-transform:rotate(90deg);-ms-transform:rotate(90deg);transform:rotate(90deg);opacity:0}}.rotateOutUpRight{-webkit-animation-name:rotateOutUpRight;animation-name:rotateOutUpRight}@-webkit-keyframes slideInDown{0%{opacity:0;-webkit-transform:translateY(-2000px);transform:translateY(-2000px)}100%{-webkit-transform:translateY(0);transform:translateY(0)}}@keyframes slideInDown{0%{opacity:0;-webkit-transform:translateY(-2000px);-ms-transform:translateY(-2000px);transform:translateY(-2000px)}100%{-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}}.slideInDown{-webkit-animation-name:slideInDown;animation-name:slideInDown}@-webkit-keyframes slideInLeft{0%{opacity:0;-webkit-transform:translateX(-2000px);transform:translateX(-2000px)}100%{-webkit-transform:translateX(0);transform:translateX(0)}}@keyframes slideInLeft{0%{opacity:0;-webkit-transform:translateX(-2000px);-ms-transform:translateX(-2000px);transform:translateX(-2000px)}100%{-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}}.slideInLeft{-webkit-animation-name:slideInLeft;animation-name:slideInLeft}@-webkit-keyframes slideInRight{0%{opacity:0;-webkit-transform:translateX(2000px);transform:translateX(2000px)}100%{-webkit-transform:translateX(0);transform:translateX(0)}}@keyframes slideInRight{0%{opacity:0;-webkit-transform:translateX(2000px);-ms-transform:translateX(2000px);transform:translateX(2000px)}100%{-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}}.slideInRight{-webkit-animation-name:slideInRight;animation-name:slideInRight}@-webkit-keyframes slideOutLeft{0%{-webkit-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(-2000px);transform:translateX(-2000px)}}@keyframes slideOutLeft{0%{-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(-2000px);-ms-transform:translateX(-2000px);transform:translateX(-2000px)}}.slideOutLeft{-webkit-animation-name:slideOutLeft;animation-name:slideOutLeft}@-webkit-keyframes slideOutRight{0%{-webkit-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(2000px);transform:translateX(2000px)}}@keyframes slideOutRight{0%{-webkit-transform:translateX(0);-ms-transform:translateX(0);transform:translateX(0)}100%{opacity:0;-webkit-transform:translateX(2000px);-ms-transform:translateX(2000px);transform:translateX(2000px)}}.slideOutRight{-webkit-animation-name:slideOutRight;animation-name:slideOutRight}@-webkit-keyframes slideOutUp{0%{-webkit-transform:translateY(0);transform:translateY(0)}100%{opacity:0;-webkit-transform:translateY(-2000px);transform:translateY(-2000px)}}@keyframes slideOutUp{0%{-webkit-transform:translateY(0);-ms-transform:translateY(0);transform:translateY(0)}100%{opacity:0;-webkit-transform:translateY(-2000px);-ms-transform:translateY(-2000px);transform:translateY(-2000px)}}.slideOutUp{-webkit-animation-name:slideOutUp;animation-name:slideOutUp}@-webkit-keyframes hinge{0%{-webkit-transform:rotate(0);transform:rotate(0);-webkit-transform-origin:top left;transform-origin:top left;-webkit-animation-timing-function:ease-in-out;animation-timing-function:ease-in-out}20%,60%{-webkit-transform:rotate(80deg);transform:rotate(80deg);-webkit-transform-origin:top left;transform-origin:top left;-webkit-animation-timing-function:ease-in-out;animation-timing-function:ease-in-out}40%{-webkit-transform:rotate(60deg);transform:rotate(60deg);-webkit-transform-origin:top left;transform-origin:top left;-webkit-animation-timing-function:ease-in-out;animation-timing-function:ease-in-out}80%{-webkit-transform:rotate(60deg) translateY(0);transform:rotate(60deg) translateY(0);opacity:1;-webkit-transform-origin:top left;transform-origin:top left;-webkit-animation-timing-function:ease-in-out;animation-timing-function:ease-in-out}100%{-webkit-transform:translateY(700px);transform:translateY(700px);opacity:0}}@keyframes hinge{0%{-webkit-transform:rotate(0);-ms-transform:rotate(0);transform:rotate(0);-webkit-transform-origin:top left;-ms-transform-origin:top left;transform-origin:top left;-webkit-animation-timing-function:ease-in-out;animation-timing-function:ease-in-out}20%,60%{-webkit-transform:rotate(80deg);-ms-transform:rotate(80deg);transform:rotate(80deg);-webkit-transform-origin:top left;-ms-transform-origin:top left;transform-origin:top left;-webkit-animation-timing-function:ease-in-out;animation-timing-function:ease-in-out}40%{-webkit-transform:rotate(60deg);-ms-transform:rotate(60deg);transform:rotate(60deg);-webkit-transform-origin:top left;-ms-transform-origin:top left;transform-origin:top left;-webkit-animation-timing-function:ease-in-out;animation-timing-function:ease-in-out}80%{-webkit-transform:rotate(60deg) translateY(0);-ms-transform:rotate(60deg) translateY(0);transform:rotate(60deg) translateY(0);opacity:1;-webkit-transform-origin:top left;-ms-transform-origin:top left;transform-origin:top left;-webkit-animation-timing-function:ease-in-out;animation-timing-function:ease-in-out}100%{-webkit-transform:translateY(700px);-ms-transform:translateY(700px);transform:translateY(700px);opacity:0}}.hinge{-webkit-animation-name:hinge;animation-name:hinge}@-webkit-keyframes rollIn{0%{opacity:0;-webkit-transform:translateX(-100%) rotate(-120deg);transform:translateX(-100%) rotate(-120deg)}100%{opacity:1;-webkit-transform:translateX(0px) rotate(0deg);transform:translateX(0px) rotate(0deg)}}@keyframes rollIn{0%{opacity:0;-webkit-transform:translateX(-100%) rotate(-120deg);-ms-transform:translateX(-100%) rotate(-120deg);transform:translateX(-100%) rotate(-120deg)}100%{opacity:1;-webkit-transform:translateX(0px) rotate(0deg);-ms-transform:translateX(0px) rotate(0deg);transform:translateX(0px) rotate(0deg)}}.rollIn{-webkit-animation-name:rollIn;animation-name:rollIn}@-webkit-keyframes rollOut{0%{opacity:1;-webkit-transform:translateX(0px) rotate(0deg);transform:translateX(0px) rotate(0deg)}100%{opacity:0;-webkit-transform:translateX(100%) rotate(120deg);transform:translateX(100%) rotate(120deg)}}@keyframes rollOut{0%{opacity:1;-webkit-transform:translateX(0px) rotate(0deg);-ms-transform:translateX(0px) rotate(0deg);transform:translateX(0px) rotate(0deg)}100%{opacity:0;-webkit-transform:translateX(100%) rotate(120deg);-ms-transform:translateX(100%) rotate(120deg);transform:translateX(100%) rotate(120deg)}}.rollOut{-webkit-animation-name:rollOut;animation-name:rollOut}


--------------------------------------------------------------------------------
/classifier.py:
--------------------------------------------------------------------------------
  1 | # -*- coding: utf-8 -*-
  2 | """
  3 | Created on Sun Nov  5 12:58:52 2017
  4 | 
  5 | @author: NishitP
  6 | """
  7 | 
  8 | import DataPrep
  9 | import FeatureSelection
 10 | import numpy as np
 11 | import pandas as pd
 12 | import pickle
 13 | from sklearn.feature_extraction.text import CountVectorizer
 14 | from sklearn.feature_extraction.text import TfidfTransformer
 15 | from sklearn.feature_extraction.text import TfidfVectorizer
 16 | from sklearn.pipeline import Pipeline
 17 | from sklearn.naive_bayes import MultinomialNB
 18 | from sklearn.linear_model import  LogisticRegression
 19 | from sklearn.linear_model import SGDClassifier
 20 | from sklearn import svm
 21 | from sklearn.ensemble import RandomForestClassifier
 22 | from sklearn.model_selection import KFold
 23 | from sklearn.metrics import confusion_matrix, f1_score, classification_report
 24 | from sklearn.model_selection import GridSearchCV
 25 | from sklearn.model_selection import learning_curve
 26 | import matplotlib.pyplot as plt
 27 | from sklearn.metrics import precision_recall_curve
 28 | from sklearn.metrics import average_precision_score
 29 | 
 30 | #string to test
 31 | doc_new = ['obama is running for president in 2016']
 32 | 
 33 | #the feature selection has been done in FeatureSelection.py module. here we will create models using those features for prediction
 34 | 
 35 | #first we will use bag of words techniques
 36 | 
 37 | #building classifier using naive bayes 
 38 | nb_pipeline = Pipeline([
 39 |         ('NBCV',FeatureSelection.countV),
 40 |         ('nb_clf',MultinomialNB())])
 41 | 
 42 | nb_pipeline.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
 43 | predicted_nb = nb_pipeline.predict(DataPrep.test_news['Statement'])
 44 | np.mean(predicted_nb == DataPrep.test_news['Label'])
 45 | 
 46 | 
 47 | #building classifier using logistic regression
 48 | logR_pipeline = Pipeline([
 49 |         ('LogRCV',FeatureSelection.countV),
 50 |         ('LogR_clf',LogisticRegression())
 51 |         ])
 52 | 
 53 | logR_pipeline.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
 54 | predicted_LogR = logR_pipeline.predict(DataPrep.test_news['Statement'])
 55 | np.mean(predicted_LogR == DataPrep.test_news['Label'])
 56 | 
 57 | 
 58 | #building Linear SVM classfier
 59 | svm_pipeline = Pipeline([
 60 |         ('svmCV',FeatureSelection.countV),
 61 |         ('svm_clf',svm.LinearSVC())
 62 |         ])
 63 | 
 64 | svm_pipeline.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
 65 | predicted_svm = svm_pipeline.predict(DataPrep.test_news['Statement'])
 66 | np.mean(predicted_svm == DataPrep.test_news['Label'])
 67 | 
 68 | 
 69 | #using SVM Stochastic Gradient Descent on hinge loss
 70 | sgd_pipeline = Pipeline([
 71 |         ('svm2CV',FeatureSelection.countV),
 72 |         ('svm2_clf',SGDClassifier(loss='hinge', penalty='l2', alpha=1e-3, n_iter=5))
 73 |         ])
 74 | 
 75 | sgd_pipeline.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
 76 | predicted_sgd = sgd_pipeline.predict(DataPrep.test_news['Statement'])
 77 | np.mean(predicted_sgd == DataPrep.test_news['Label'])
 78 | 
 79 | 
 80 | #random forest
 81 | random_forest = Pipeline([
 82 |         ('rfCV',FeatureSelection.countV),
 83 |         ('rf_clf',RandomForestClassifier(n_estimators=200,n_jobs=3))
 84 |         ])
 85 |     
 86 | random_forest.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
 87 | predicted_rf = random_forest.predict(DataPrep.test_news['Statement'])
 88 | np.mean(predicted_rf == DataPrep.test_news['Label'])
 89 | 
 90 | 
 91 | #User defined functon for K-Fold cross validatoin
 92 | def build_confusion_matrix(classifier):
 93 |     
 94 |     k_fold = KFold(n_splits=5)
 95 |     scores = []
 96 |     confusion = np.array([[0,0],[0,0]])
 97 | 
 98 |     for train_ind, test_ind in k_fold.split(DataPrep.train_news):
 99 |         train_text = DataPrep.train_news.iloc[train_ind]['Statement'] 
100 |         train_y = DataPrep.train_news.iloc[train_ind]['Label']
101 |     
102 |         test_text = DataPrep.train_news.iloc[test_ind]['Statement']
103 |         test_y = DataPrep.train_news.iloc[test_ind]['Label']
104 |         
105 |         classifier.fit(train_text,train_y)
106 |         predictions = classifier.predict(test_text)
107 |         
108 |         confusion += confusion_matrix(test_y,predictions)
109 |         score = f1_score(test_y,predictions)
110 |         scores.append(score)
111 |     
112 |     return (print('Total statements classified:', len(DataPrep.train_news)),
113 |     print('Score:', sum(scores)/len(scores)),
114 |     print('score length', len(scores)),
115 |     print('Confusion matrix:'),
116 |     print(confusion))
117 |     
118 | #K-fold cross validation for all classifiers
119 | build_confusion_matrix(nb_pipeline)
120 | build_confusion_matrix(logR_pipeline)
121 | build_confusion_matrix(svm_pipeline)
122 | build_confusion_matrix(sgd_pipeline)
123 | build_confusion_matrix(random_forest)
124 | 
125 | #========================================================================================
126 | #Bag of words confusion matrix and F1 scores
127 | 
128 | #Naive bayes
129 | # [2118 2370]
130 | # [1664 4088]
131 | # f1-Score: 0.669611539651
132 | 
133 | #Logistic regression
134 | # [2252 2236]
135 | # [1933 3819]
136 | # f1-Score: 0.646909097798
137 | 
138 | #svm
139 | # [2260 2228]
140 | # [2246 3506]
141 | #f1-score: 0.610468748792
142 | 
143 | #sgdclassifier
144 | # [2414 2074]
145 | # [2042 3710]
146 | # f1-Score: 0.640874558778
147 | 
148 | #random forest classifier
149 | # [1821 2667]
150 | # [1192 4560]
151 | # f1-Score: 0.702651511011
152 | #=========================================================================================
153 | 
154 | 
155 | """So far we have used bag of words technique to extract the features and passed those featuers into classifiers. We have also seen the
156 | f1 scores of these classifiers. now lets enhance these features using term frequency weights with various n-grams
157 | """
158 | 
159 | ##Now using n-grams
160 | #naive-bayes classifier
161 | nb_pipeline_ngram = Pipeline([
162 |         ('nb_tfidf',FeatureSelection.tfidf_ngram),
163 |         ('nb_clf',MultinomialNB())])
164 | 
165 | nb_pipeline_ngram.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
166 | predicted_nb_ngram = nb_pipeline_ngram.predict(DataPrep.test_news['Statement'])
167 | np.mean(predicted_nb_ngram == DataPrep.test_news['Label'])
168 | 
169 | 
170 | #logistic regression classifier
171 | logR_pipeline_ngram = Pipeline([
172 |         ('LogR_tfidf',FeatureSelection.tfidf_ngram),
173 |         ('LogR_clf',LogisticRegression(penalty="l2",C=1))
174 |         ])
175 | 
176 | logR_pipeline_ngram.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
177 | predicted_LogR_ngram = logR_pipeline_ngram.predict(DataPrep.test_news['Statement'])
178 | np.mean(predicted_LogR_ngram == DataPrep.test_news['Label'])
179 | 
180 | 
181 | #linear SVM classifier
182 | svm_pipeline_ngram = Pipeline([
183 |         ('svm_tfidf',FeatureSelection.tfidf_ngram),
184 |         ('svm_clf',svm.LinearSVC())
185 |         ])
186 | 
187 | svm_pipeline_ngram.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
188 | predicted_svm_ngram = svm_pipeline_ngram.predict(DataPrep.test_news['Statement'])
189 | np.mean(predicted_svm_ngram == DataPrep.test_news['Label'])
190 | 
191 | 
192 | #sgd classifier
193 | sgd_pipeline_ngram = Pipeline([
194 |          ('sgd_tfidf',FeatureSelection.tfidf_ngram),
195 |          ('sgd_clf',SGDClassifier(loss='hinge', penalty='l2', alpha=1e-3, n_iter=5))
196 |          ])
197 | 
198 | sgd_pipeline_ngram.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
199 | predicted_sgd_ngram = sgd_pipeline_ngram.predict(DataPrep.test_news['Statement'])
200 | np.mean(predicted_sgd_ngram == DataPrep.test_news['Label'])
201 | 
202 | 
203 | #random forest classifier
204 | random_forest_ngram = Pipeline([
205 |         ('rf_tfidf',FeatureSelection.tfidf_ngram),
206 |         ('rf_clf',RandomForestClassifier(n_estimators=300,n_jobs=3))
207 |         ])
208 |     
209 | random_forest_ngram.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
210 | predicted_rf_ngram = random_forest_ngram.predict(DataPrep.test_news['Statement'])
211 | np.mean(predicted_rf_ngram == DataPrep.test_news['Label'])
212 | 
213 | 
214 | #K-fold cross validation for all classifiers
215 | build_confusion_matrix(nb_pipeline_ngram)
216 | build_confusion_matrix(logR_pipeline_ngram)
217 | build_confusion_matrix(svm_pipeline_ngram)
218 | build_confusion_matrix(sgd_pipeline_ngram)
219 | build_confusion_matrix(random_forest_ngram)
220 | 
221 | #========================================================================================
222 | #n-grams & tfidf confusion matrix and F1 scores
223 | 
224 | #Naive bayes
225 | # [841 3647]
226 | # [427 5325]
227 | # f1-Score: 0.723262051071
228 | 
229 | #Logistic regression
230 | # [1617 2871]
231 | # [1097 4655]
232 | # f1-Score: 0.70113000531
233 | 
234 | #svm
235 | # [2016 2472]
236 | # [1524 4228]
237 | # f1-Score: 0.67909201429
238 | 
239 | #sgdclassifier
240 | # [  10 4478]
241 | # [  13 5739]
242 | # f1-Score: 0.718731637053
243 | 
244 | #random forest
245 | # [1979 2509]
246 | # [1630 4122]
247 | # f1-Score: 0.665720333284
248 | #=========================================================================================
249 | 
250 | print(classification_report(DataPrep.test_news['Label'], predicted_nb_ngram))
251 | print(classification_report(DataPrep.test_news['Label'], predicted_LogR_ngram))
252 | print(classification_report(DataPrep.test_news['Label'], predicted_svm_ngram))
253 | print(classification_report(DataPrep.test_news['Label'], predicted_sgd_ngram))
254 | print(classification_report(DataPrep.test_news['Label'], predicted_rf_ngram))
255 | 
256 | DataPrep.test_news['Label'].shape
257 | 
258 | """
259 | Out of all the models fitted, we would take 2 best performing model. we would call them candidate models
260 | from the confusion matrix, we can see that random forest and logistic regression are best performing 
261 | in terms of precision and recall (take a look into false positive and true negative counts which appeares
262 | to be low compared to rest of the models)
263 | """
264 | 
265 | #grid-search parameter optimization
266 | #random forest classifier parameters
267 | parameters = {'rf_tfidf__ngram_range': [(1, 1), (1, 2),(1,3),(1,4),(1,5)],
268 |                'rf_tfidf__use_idf': (True, False),
269 |                'rf_clf__max_depth': (1,2,3,4,5,6,7,8,9,10,11,12,13,14,15)
270 | }
271 | 
272 | gs_clf = GridSearchCV(random_forest_ngram, parameters, n_jobs=-1)
273 | gs_clf = gs_clf.fit(DataPrep.train_news['Statement'][:10000],DataPrep.train_news['Label'][:10000])
274 | 
275 | gs_clf.best_score_
276 | gs_clf.best_params_
277 | gs_clf.cv_results_
278 | 
279 | #logistic regression parameters
280 | parameters = {'LogR_tfidf__ngram_range': [(1, 1), (1, 2),(1,3),(1,4),(1,5)],
281 |                'LogR_tfidf__use_idf': (True, False),
282 |                'LogR_tfidf__smooth_idf': (True, False)
283 | }
284 | 
285 | gs_clf = GridSearchCV(logR_pipeline_ngram, parameters, n_jobs=-1)
286 | gs_clf = gs_clf.fit(DataPrep.train_news['Statement'][:10000],DataPrep.train_news['Label'][:10000])
287 | 
288 | gs_clf.best_score_
289 | gs_clf.best_params_
290 | gs_clf.cv_results_
291 | 
292 | #Linear SVM 
293 | parameters = {'svm_tfidf__ngram_range': [(1, 1), (1, 2),(1,3),(1,4),(1,5)],
294 |                'svm_tfidf__use_idf': (True, False),
295 |                'svm_tfidf__smooth_idf': (True, False),
296 |                'svm_clf__penalty': ('l1','l2'),
297 | }
298 | 
299 | gs_clf = GridSearchCV(svm_pipeline_ngram, parameters, n_jobs=-1)
300 | gs_clf = gs_clf.fit(DataPrep.train_news['Statement'][:10000],DataPrep.train_news['Label'][:10000])
301 | 
302 | gs_clf.best_score_
303 | gs_clf.best_params_
304 | gs_clf.cv_results_
305 | 
306 | #by running above commands we can find the model with best performing parameters
307 | 
308 | 
309 | #running both random forest and logistic regression models again with best parameter found with GridSearch method
310 | random_forest_final = Pipeline([
311 |         ('rf_tfidf',TfidfVectorizer(stop_words='english',ngram_range=(1,3),use_idf=True,smooth_idf=True)),
312 |         ('rf_clf',RandomForestClassifier(n_estimators=300,n_jobs=3,max_depth=10))
313 |         ])
314 |     
315 | random_forest_final.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
316 | predicted_rf_final = random_forest_final.predict(DataPrep.test_news['Statement'])
317 | np.mean(predicted_rf_final == DataPrep.test_news['Label'])
318 | print(metrics.classification_report(DataPrep.test_news['Label'], predicted_rf_final))
319 | 
320 | logR_pipeline_final = Pipeline([
321 |         #('LogRCV',countV_ngram),
322 |         ('LogR_tfidf',TfidfVectorizer(stop_words='english',ngram_range=(1,5),use_idf=True,smooth_idf=False)),
323 |         ('LogR_clf',LogisticRegression(penalty="l2",C=1))
324 |         ])
325 | 
326 | logR_pipeline_final.fit(DataPrep.train_news['Statement'],DataPrep.train_news['Label'])
327 | predicted_LogR_final = logR_pipeline_final.predict(DataPrep.test_news['Statement'])
328 | np.mean(predicted_LogR_final == DataPrep.test_news['Label'])
329 | #accuracy = 0.62
330 | print(metrics.classification_report(DataPrep.test_news['Label'], predicted_LogR_final))
331 | 
332 | 
333 | """
334 | by running both random forest and logistic regression with GridSearch's best parameter estimation, we found that for random 
335 | forest model with n-gram has better accuracty than with the parameter estimated. The logistic regression model with best parameter 
336 | has almost similar performance as n-gram model so logistic regression will be out choice of model for prediction.
337 | """
338 | 
339 | #saving best model to the disk
340 | model_file = 'final_model.sav'
341 | pickle.dump(logR_pipeline_ngram,open(model_file,'wb'))
342 | 
343 | 
344 | #Plotting learing curve
345 | def plot_learing_curve(pipeline,title):
346 |     size = 10000
347 |     cv = KFold(size, shuffle=True)
348 |     
349 |     X = DataPrep.train_news["Statement"]
350 |     y = DataPrep.train_news["Label"]
351 |     
352 |     pl = pipeline
353 |     pl.fit(X,y)
354 |     
355 |     train_sizes, train_scores, test_scores = learning_curve(pl, X, y, n_jobs=-1, cv=cv, train_sizes=np.linspace(.1, 1.0, 5), verbose=0)
356 |        
357 |     train_scores_mean = np.mean(train_scores, axis=1)
358 |     train_scores_std = np.std(train_scores, axis=1)
359 |     test_scores_mean = np.mean(test_scores, axis=1)
360 |     test_scores_std = np.std(test_scores, axis=1)
361 |      
362 |     plt.figure()
363 |     plt.title(title)
364 |     plt.legend(loc="best")
365 |     plt.xlabel("Training examples")
366 |     plt.ylabel("Score")
367 |     plt.gca().invert_yaxis()
368 |     
369 |     # box-like grid
370 |     plt.grid()
371 |     
372 |     # plot the std deviation as a transparent range at each training set size
373 |     plt.fill_between(train_sizes, train_scores_mean - train_scores_std, train_scores_mean + train_scores_std, alpha=0.1, color="r")
374 |     plt.fill_between(train_sizes, test_scores_mean - test_scores_std, test_scores_mean + test_scores_std, alpha=0.1, color="g")
375 |     
376 |     # plot the average training and test score lines at each training set size
377 |     plt.plot(train_sizes, train_scores_mean, 'o-', color="r", label="Training score")
378 |     plt.plot(train_sizes, test_scores_mean, 'o-', color="g", label="Cross-validation score")
379 |     
380 |     # sizes the window for readability and displays the plot
381 |     # shows error from 0 to 1.1
382 |     plt.ylim(-.1,1.1)
383 |     plt.show()
384 | 
385 | 
386 | #below command will plot learing curves for each of the classifiers
387 | plot_learing_curve(logR_pipeline_ngram,"Naive-bayes Classifier")
388 | plot_learing_curve(nb_pipeline_ngram,"LogisticRegression Classifier")
389 | plot_learing_curve(svm_pipeline_ngram,"SVM Classifier")
390 | plot_learing_curve(sgd_pipeline_ngram,"SGD Classifier")
391 | plot_learing_curve(random_forest_ngram,"RandomForest Classifier")
392 | 
393 | """
394 | by plotting the learning cureve for logistic regression, it can be seen that cross-validation score is stagnating throughout and it 
395 | is unable to learn from data. Also we see that there are high errors that indicates model is simple and we may want to increase the
396 | model complexity.
397 | """
398 | 
399 | 
400 | #plotting Precision-Recall curve
401 | def plot_PR_curve(classifier):
402 |     
403 |     precision, recall, thresholds = precision_recall_curve(DataPrep.test_news['Label'], classifier)
404 |     average_precision = average_precision_score(DataPrep.test_news['Label'], classifier)
405 |     
406 |     plt.step(recall, precision, color='b', alpha=0.2,
407 |              where='post')
408 |     plt.fill_between(recall, precision, step='post', alpha=0.2,
409 |                      color='b')
410 |     
411 |     plt.xlabel('Recall')
412 |     plt.ylabel('Precision')
413 |     plt.ylim([0.0, 1.05])
414 |     plt.xlim([0.0, 1.0])
415 |     plt.title('2-class Random Forest Precision-Recall curve: AP={0:0.2f}'.format(
416 |               average_precision))
417 |     
418 | plot_PR_curve(predicted_LogR_ngram)
419 | plot_PR_curve(predicted_rf_ngram)
420 | 
421 | 
422 | """
423 | Now let's extract the most informative feature from ifidf vectorizer for all fo the classifiers and see of there are any common
424 | words that we can identify i.e. are these most informative feature acorss the classifiers are same? we will create a function that 
425 | will extract top 50 features.
426 | """
427 | 
428 | def show_most_informative_features(model, vect, clf, text=None, n=50):
429 |     # Extract the vectorizer and the classifier from the pipeline
430 |     vectorizer = model.named_steps[vect]
431 |     classifier = model.named_steps[clf]
432 | 
433 |      # Check to make sure that we can perform this computation
434 |     if not hasattr(classifier, 'coef_'):
435 |         raise TypeError(
436 |             "Cannot compute most informative features on {}.".format(
437 |                 classifier.__class__.__name__
438 |             )
439 |         )
440 |             
441 |     if text is not None:
442 |         # Compute the coefficients for the text
443 |         tvec = model.transform([text]).toarray()
444 |     else:
445 |         # Otherwise simply use the coefficients
446 |         tvec = classifier.coef_
447 | 
448 |     # Zip the feature names with the coefs and sort
449 |     coefs = sorted(
450 |         zip(tvec[0], vectorizer.get_feature_names()),
451 |         reverse=True
452 |     )
453 |     
454 |     # Get the top n and bottom n coef, name pairs
455 |     topn  = zip(coefs[:n], coefs[:-(n+1):-1])
456 | 
457 |     # Create the output string to return
458 |     output = []
459 | 
460 |     # If text, add the predicted value to the output.
461 |     if text is not None:
462 |         output.append("\"{}\"".format(text))
463 |         output.append(
464 |             "Classified as: {}".format(model.predict([text]))
465 |         )
466 |         output.append("")
467 | 
468 |     # Create two columns with most negative and most positive features.
469 |     for (cp, fnp), (cn, fnn) in topn:
470 |         output.append(
471 |             "{:0.4f}{: >15}    {:0.4f}{: >15}".format(
472 |                 cp, fnp, cn, fnn
473 |             )
474 |         )
475 |     #return "\n".join(output)
476 |     print(output)
477 | 
478 | show_most_informative_features(logR_pipeline_ngram,vect='LogR_tfidf',clf='LogR_clf')
479 | show_most_informative_features(nb_pipeline_ngram,vect='nb_tfidf',clf='nb_clf')
480 | show_most_informative_features(svm_pipeline_ngram,vect='svm_tfidf',clf='svm_clf')
481 | show_most_informative_features(sgd_pipeline_ngram,vect='sgd_tfidf',clf='sgd_clf')
482 | 


--------------------------------------------------------------------------------
/final-fnd.ipynb:
--------------------------------------------------------------------------------
   1 | {
   2 |  "cells": [
   3 |   {
   4 |    "cell_type": "code",
   5 |    "execution_count": 45,
   6 |    "metadata": {
   7 |     "execution": {
   8 |      "iopub.execute_input": "2021-05-25T06:50:29.636394Z",
   9 |      "iopub.status.busy": "2021-05-25T06:50:29.636041Z",
  10 |      "iopub.status.idle": "2021-05-25T06:50:29.643277Z",
  11 |      "shell.execute_reply": "2021-05-25T06:50:29.642127Z",
  12 |      "shell.execute_reply.started": "2021-05-25T06:50:29.636365Z"
  13 |     }
  14 |    },
  15 |    "outputs": [],
  16 |    "source": [
  17 |     "import pandas as pd\n",
  18 |     "import sklearn\n",
  19 |     "import itertools\n",
  20 |     "import numpy as np\n",
  21 |     "import seaborn as sb\n",
  22 |     "import re\n",
  23 |     "import nltk\n",
  24 |     "import pickle\n",
  25 |     "from sklearn.model_selection import train_test_split\n",
  26 |     "from sklearn.feature_extraction.text import TfidfVectorizer\n",
  27 |     "from sklearn import metrics\n",
  28 |     "from sklearn.metrics import confusion_matrix\n",
  29 |     "from matplotlib import pyplot as plt\n",
  30 |     "from sklearn.linear_model import PassiveAggressiveClassifier\n",
  31 |     "from nltk.stem import WordNetLemmatizer\n",
  32 |     "from nltk.corpus import stopwords"
  33 |    ]
  34 |   },
  35 |   {
  36 |    "cell_type": "code",
  37 |    "execution_count": 46,
  38 |    "metadata": {
  39 |     "execution": {
  40 |      "iopub.execute_input": "2021-05-25T06:50:29.656569Z",
  41 |      "iopub.status.busy": "2021-05-25T06:50:29.656203Z",
  42 |      "iopub.status.idle": "2021-05-25T06:50:32.048864Z",
  43 |      "shell.execute_reply": "2021-05-25T06:50:32.047882Z",
  44 |      "shell.execute_reply.started": "2021-05-25T06:50:29.65654Z"
  45 |     }
  46 |    },
  47 |    "outputs": [],
  48 |    "source": [
  49 |     "train_df = pd.read_csv(r'C:\\Users\\Mayur\\Downloads\\train.csv')"
  50 |    ]
  51 |   },
  52 |   {
  53 |    "cell_type": "code",
  54 |    "execution_count": 47,
  55 |    "metadata": {
  56 |     "execution": {
  57 |      "iopub.execute_input": "2021-05-25T06:50:32.05136Z",
  58 |      "iopub.status.busy": "2021-05-25T06:50:32.051032Z",
  59 |      "iopub.status.idle": "2021-05-25T06:50:32.089516Z",
  60 |      "shell.execute_reply": "2021-05-25T06:50:32.088399Z",
  61 |      "shell.execute_reply.started": "2021-05-25T06:50:32.051329Z"
  62 |     }
  63 |    },
  64 |    "outputs": [
  65 |     {
  66 |      "data": {
  67 |       "text/html": [
  68 |        "<div>\n",
  69 |        "<style scoped>\n",
  70 |        "    .dataframe tbody tr th:only-of-type {\n",
  71 |        "        vertical-align: middle;\n",
  72 |        "    }\n",
  73 |        "\n",
  74 |        "    .dataframe tbody tr th {\n",
  75 |        "        vertical-align: top;\n",
  76 |        "    }\n",
  77 |        "\n",
  78 |        "    .dataframe thead th {\n",
  79 |        "        text-align: right;\n",
  80 |        "    }\n",
  81 |        "</style>\n",
  82 |        "<table border=\"1\" class=\"dataframe\">\n",
  83 |        "  <thead>\n",
  84 |        "    <tr style=\"text-align: right;\">\n",
  85 |        "      <th></th>\n",
  86 |        "      <th>id</th>\n",
  87 |        "      <th>title</th>\n",
  88 |        "      <th>author</th>\n",
  89 |        "      <th>text</th>\n",
  90 |        "      <th>label</th>\n",
  91 |        "    </tr>\n",
  92 |        "  </thead>\n",
  93 |        "  <tbody>\n",
  94 |        "    <tr>\n",
  95 |        "      <th>0</th>\n",
  96 |        "      <td>0</td>\n",
  97 |        "      <td>House Dem Aide: We Didn’t Even See Comey’s Let...</td>\n",
  98 |        "      <td>Darrell Lucus</td>\n",
  99 |        "      <td>House Dem Aide: We Didn’t Even See Comey’s Let...</td>\n",
 100 |        "      <td>1</td>\n",
 101 |        "    </tr>\n",
 102 |        "    <tr>\n",
 103 |        "      <th>1</th>\n",
 104 |        "      <td>1</td>\n",
 105 |        "      <td>FLYNN: Hillary Clinton, Big Woman on Campus - ...</td>\n",
 106 |        "      <td>Daniel J. Flynn</td>\n",
 107 |        "      <td>Ever get the feeling your life circles the rou...</td>\n",
 108 |        "      <td>0</td>\n",
 109 |        "    </tr>\n",
 110 |        "    <tr>\n",
 111 |        "      <th>2</th>\n",
 112 |        "      <td>2</td>\n",
 113 |        "      <td>Why the Truth Might Get You Fired</td>\n",
 114 |        "      <td>Consortiumnews.com</td>\n",
 115 |        "      <td>Why the Truth Might Get You Fired October 29, ...</td>\n",
 116 |        "      <td>1</td>\n",
 117 |        "    </tr>\n",
 118 |        "    <tr>\n",
 119 |        "      <th>3</th>\n",
 120 |        "      <td>3</td>\n",
 121 |        "      <td>15 Civilians Killed In Single US Airstrike Hav...</td>\n",
 122 |        "      <td>Jessica Purkiss</td>\n",
 123 |        "      <td>Videos 15 Civilians Killed In Single US Airstr...</td>\n",
 124 |        "      <td>1</td>\n",
 125 |        "    </tr>\n",
 126 |        "    <tr>\n",
 127 |        "      <th>4</th>\n",
 128 |        "      <td>4</td>\n",
 129 |        "      <td>Iranian woman jailed for fictional unpublished...</td>\n",
 130 |        "      <td>Howard Portnoy</td>\n",
 131 |        "      <td>Print \\nAn Iranian woman has been sentenced to...</td>\n",
 132 |        "      <td>1</td>\n",
 133 |        "    </tr>\n",
 134 |        "    <tr>\n",
 135 |        "      <th>5</th>\n",
 136 |        "      <td>5</td>\n",
 137 |        "      <td>Jackie Mason: Hollywood Would Love Trump if He...</td>\n",
 138 |        "      <td>Daniel Nussbaum</td>\n",
 139 |        "      <td>In these trying times, Jackie Mason is the Voi...</td>\n",
 140 |        "      <td>0</td>\n",
 141 |        "    </tr>\n",
 142 |        "    <tr>\n",
 143 |        "      <th>6</th>\n",
 144 |        "      <td>6</td>\n",
 145 |        "      <td>Life: Life Of Luxury: Elton John’s 6 Favorite ...</td>\n",
 146 |        "      <td>NaN</td>\n",
 147 |        "      <td>Ever wonder how Britain’s most iconic pop pian...</td>\n",
 148 |        "      <td>1</td>\n",
 149 |        "    </tr>\n",
 150 |        "    <tr>\n",
 151 |        "      <th>7</th>\n",
 152 |        "      <td>7</td>\n",
 153 |        "      <td>Benoît Hamon Wins French Socialist Party’s Pre...</td>\n",
 154 |        "      <td>Alissa J. Rubin</td>\n",
 155 |        "      <td>PARIS  —   France chose an idealistic, traditi...</td>\n",
 156 |        "      <td>0</td>\n",
 157 |        "    </tr>\n",
 158 |        "    <tr>\n",
 159 |        "      <th>8</th>\n",
 160 |        "      <td>8</td>\n",
 161 |        "      <td>Excerpts From a Draft Script for Donald Trump’...</td>\n",
 162 |        "      <td>NaN</td>\n",
 163 |        "      <td>Donald J. Trump is scheduled to make a highly ...</td>\n",
 164 |        "      <td>0</td>\n",
 165 |        "    </tr>\n",
 166 |        "    <tr>\n",
 167 |        "      <th>9</th>\n",
 168 |        "      <td>9</td>\n",
 169 |        "      <td>A Back-Channel Plan for Ukraine and Russia, Co...</td>\n",
 170 |        "      <td>Megan Twohey and Scott Shane</td>\n",
 171 |        "      <td>A week before Michael T. Flynn resigned as nat...</td>\n",
 172 |        "      <td>0</td>\n",
 173 |        "    </tr>\n",
 174 |        "    <tr>\n",
 175 |        "      <th>10</th>\n",
 176 |        "      <td>10</td>\n",
 177 |        "      <td>Obama’s Organizing for Action Partners with So...</td>\n",
 178 |        "      <td>Aaron Klein</td>\n",
 179 |        "      <td>Organizing for Action, the activist group that...</td>\n",
 180 |        "      <td>0</td>\n",
 181 |        "    </tr>\n",
 182 |        "    <tr>\n",
 183 |        "      <th>11</th>\n",
 184 |        "      <td>11</td>\n",
 185 |        "      <td>BBC Comedy Sketch \"Real Housewives of ISIS\" Ca...</td>\n",
 186 |        "      <td>Chris Tomlinson</td>\n",
 187 |        "      <td>The BBC produced spoof on the “Real Housewives...</td>\n",
 188 |        "      <td>0</td>\n",
 189 |        "    </tr>\n",
 190 |        "    <tr>\n",
 191 |        "      <th>12</th>\n",
 192 |        "      <td>12</td>\n",
 193 |        "      <td>Russian Researchers Discover Secret Nazi Milit...</td>\n",
 194 |        "      <td>Amando Flavio</td>\n",
 195 |        "      <td>The mystery surrounding The Third Reich and Na...</td>\n",
 196 |        "      <td>1</td>\n",
 197 |        "    </tr>\n",
 198 |        "    <tr>\n",
 199 |        "      <th>13</th>\n",
 200 |        "      <td>13</td>\n",
 201 |        "      <td>US Officials See No Link Between Trump and Russia</td>\n",
 202 |        "      <td>Jason Ditz</td>\n",
 203 |        "      <td>Clinton Campaign Demands FBI Affirm Trump's Ru...</td>\n",
 204 |        "      <td>1</td>\n",
 205 |        "    </tr>\n",
 206 |        "    <tr>\n",
 207 |        "      <th>14</th>\n",
 208 |        "      <td>14</td>\n",
 209 |        "      <td>Re: Yes, There Are Paid Government Trolls On S...</td>\n",
 210 |        "      <td>AnotherAnnie</td>\n",
 211 |        "      <td>Yes, There Are Paid Government Trolls On Socia...</td>\n",
 212 |        "      <td>1</td>\n",
 213 |        "    </tr>\n",
 214 |        "  </tbody>\n",
 215 |        "</table>\n",
 216 |        "</div>"
 217 |       ],
 218 |       "text/plain": [
 219 |        "    id                                              title  \\\n",
 220 |        "0    0  House Dem Aide: We Didn’t Even See Comey’s Let...   \n",
 221 |        "1    1  FLYNN: Hillary Clinton, Big Woman on Campus - ...   \n",
 222 |        "2    2                  Why the Truth Might Get You Fired   \n",
 223 |        "3    3  15 Civilians Killed In Single US Airstrike Hav...   \n",
 224 |        "4    4  Iranian woman jailed for fictional unpublished...   \n",
 225 |        "5    5  Jackie Mason: Hollywood Would Love Trump if He...   \n",
 226 |        "6    6  Life: Life Of Luxury: Elton John’s 6 Favorite ...   \n",
 227 |        "7    7  Benoît Hamon Wins French Socialist Party’s Pre...   \n",
 228 |        "8    8  Excerpts From a Draft Script for Donald Trump’...   \n",
 229 |        "9    9  A Back-Channel Plan for Ukraine and Russia, Co...   \n",
 230 |        "10  10  Obama’s Organizing for Action Partners with So...   \n",
 231 |        "11  11  BBC Comedy Sketch \"Real Housewives of ISIS\" Ca...   \n",
 232 |        "12  12  Russian Researchers Discover Secret Nazi Milit...   \n",
 233 |        "13  13  US Officials See No Link Between Trump and Russia   \n",
 234 |        "14  14  Re: Yes, There Are Paid Government Trolls On S...   \n",
 235 |        "\n",
 236 |        "                          author  \\\n",
 237 |        "0                  Darrell Lucus   \n",
 238 |        "1                Daniel J. Flynn   \n",
 239 |        "2             Consortiumnews.com   \n",
 240 |        "3                Jessica Purkiss   \n",
 241 |        "4                 Howard Portnoy   \n",
 242 |        "5                Daniel Nussbaum   \n",
 243 |        "6                            NaN   \n",
 244 |        "7                Alissa J. Rubin   \n",
 245 |        "8                            NaN   \n",
 246 |        "9   Megan Twohey and Scott Shane   \n",
 247 |        "10                   Aaron Klein   \n",
 248 |        "11               Chris Tomlinson   \n",
 249 |        "12                 Amando Flavio   \n",
 250 |        "13                    Jason Ditz   \n",
 251 |        "14                  AnotherAnnie   \n",
 252 |        "\n",
 253 |        "                                                 text  label  \n",
 254 |        "0   House Dem Aide: We Didn’t Even See Comey’s Let...      1  \n",
 255 |        "1   Ever get the feeling your life circles the rou...      0  \n",
 256 |        "2   Why the Truth Might Get You Fired October 29, ...      1  \n",
 257 |        "3   Videos 15 Civilians Killed In Single US Airstr...      1  \n",
 258 |        "4   Print \\nAn Iranian woman has been sentenced to...      1  \n",
 259 |        "5   In these trying times, Jackie Mason is the Voi...      0  \n",
 260 |        "6   Ever wonder how Britain’s most iconic pop pian...      1  \n",
 261 |        "7   PARIS  —   France chose an idealistic, traditi...      0  \n",
 262 |        "8   Donald J. Trump is scheduled to make a highly ...      0  \n",
 263 |        "9   A week before Michael T. Flynn resigned as nat...      0  \n",
 264 |        "10  Organizing for Action, the activist group that...      0  \n",
 265 |        "11  The BBC produced spoof on the “Real Housewives...      0  \n",
 266 |        "12  The mystery surrounding The Third Reich and Na...      1  \n",
 267 |        "13  Clinton Campaign Demands FBI Affirm Trump's Ru...      1  \n",
 268 |        "14  Yes, There Are Paid Government Trolls On Socia...      1  "
 269 |       ]
 270 |      },
 271 |      "execution_count": 47,
 272 |      "metadata": {},
 273 |      "output_type": "execute_result"
 274 |     }
 275 |    ],
 276 |    "source": [
 277 |     "train_df.head(15)"
 278 |    ]
 279 |   },
 280 |   {
 281 |    "cell_type": "code",
 282 |    "execution_count": 48,
 283 |    "metadata": {
 284 |     "execution": {
 285 |      "iopub.execute_input": "2021-05-25T06:50:32.091153Z",
 286 |      "iopub.status.busy": "2021-05-25T06:50:32.090847Z",
 287 |      "iopub.status.idle": "2021-05-25T06:50:32.104756Z",
 288 |      "shell.execute_reply": "2021-05-25T06:50:32.10363Z",
 289 |      "shell.execute_reply.started": "2021-05-25T06:50:32.091126Z"
 290 |     }
 291 |    },
 292 |    "outputs": [],
 293 |    "source": [
 294 |     "train_df = train_df.drop(\"author\", axis = 1)\n",
 295 |     "train_df = train_df.drop(\"title\", axis = 1)\n",
 296 |     "train_df = train_df.drop(\"id\", axis = 1)"
 297 |    ]
 298 |   },
 299 |   {
 300 |    "cell_type": "code",
 301 |    "execution_count": 49,
 302 |    "metadata": {
 303 |     "execution": {
 304 |      "iopub.execute_input": "2021-05-25T06:50:32.10674Z",
 305 |      "iopub.status.busy": "2021-05-25T06:50:32.106434Z",
 306 |      "iopub.status.idle": "2021-05-25T06:50:32.120541Z",
 307 |      "shell.execute_reply": "2021-05-25T06:50:32.119386Z",
 308 |      "shell.execute_reply.started": "2021-05-25T06:50:32.106712Z"
 309 |     }
 310 |    },
 311 |    "outputs": [
 312 |     {
 313 |      "data": {
 314 |       "text/plain": [
 315 |        "(20800, 2)"
 316 |       ]
 317 |      },
 318 |      "execution_count": 49,
 319 |      "metadata": {},
 320 |      "output_type": "execute_result"
 321 |     }
 322 |    ],
 323 |    "source": [
 324 |     "train_df.shape"
 325 |    ]
 326 |   },
 327 |   {
 328 |    "cell_type": "code",
 329 |    "execution_count": 50,
 330 |    "metadata": {
 331 |     "execution": {
 332 |      "iopub.execute_input": "2021-05-25T06:50:32.124489Z",
 333 |      "iopub.status.busy": "2021-05-25T06:50:32.12414Z",
 334 |      "iopub.status.idle": "2021-05-25T06:50:32.140229Z",
 335 |      "shell.execute_reply": "2021-05-25T06:50:32.139288Z",
 336 |      "shell.execute_reply.started": "2021-05-25T06:50:32.124461Z"
 337 |     }
 338 |    },
 339 |    "outputs": [
 340 |     {
 341 |      "data": {
 342 |       "text/html": [
 343 |        "<div>\n",
 344 |        "<style scoped>\n",
 345 |        "    .dataframe tbody tr th:only-of-type {\n",
 346 |        "        vertical-align: middle;\n",
 347 |        "    }\n",
 348 |        "\n",
 349 |        "    .dataframe tbody tr th {\n",
 350 |        "        vertical-align: top;\n",
 351 |        "    }\n",
 352 |        "\n",
 353 |        "    .dataframe thead th {\n",
 354 |        "        text-align: right;\n",
 355 |        "    }\n",
 356 |        "</style>\n",
 357 |        "<table border=\"1\" class=\"dataframe\">\n",
 358 |        "  <thead>\n",
 359 |        "    <tr style=\"text-align: right;\">\n",
 360 |        "      <th></th>\n",
 361 |        "      <th>text</th>\n",
 362 |        "      <th>label</th>\n",
 363 |        "    </tr>\n",
 364 |        "  </thead>\n",
 365 |        "  <tbody>\n",
 366 |        "    <tr>\n",
 367 |        "      <th>0</th>\n",
 368 |        "      <td>House Dem Aide: We Didn’t Even See Comey’s Let...</td>\n",
 369 |        "      <td>1</td>\n",
 370 |        "    </tr>\n",
 371 |        "    <tr>\n",
 372 |        "      <th>1</th>\n",
 373 |        "      <td>Ever get the feeling your life circles the rou...</td>\n",
 374 |        "      <td>0</td>\n",
 375 |        "    </tr>\n",
 376 |        "    <tr>\n",
 377 |        "      <th>2</th>\n",
 378 |        "      <td>Why the Truth Might Get You Fired October 29, ...</td>\n",
 379 |        "      <td>1</td>\n",
 380 |        "    </tr>\n",
 381 |        "    <tr>\n",
 382 |        "      <th>3</th>\n",
 383 |        "      <td>Videos 15 Civilians Killed In Single US Airstr...</td>\n",
 384 |        "      <td>1</td>\n",
 385 |        "    </tr>\n",
 386 |        "    <tr>\n",
 387 |        "      <th>4</th>\n",
 388 |        "      <td>Print \\nAn Iranian woman has been sentenced to...</td>\n",
 389 |        "      <td>1</td>\n",
 390 |        "    </tr>\n",
 391 |        "    <tr>\n",
 392 |        "      <th>5</th>\n",
 393 |        "      <td>In these trying times, Jackie Mason is the Voi...</td>\n",
 394 |        "      <td>0</td>\n",
 395 |        "    </tr>\n",
 396 |        "    <tr>\n",
 397 |        "      <th>6</th>\n",
 398 |        "      <td>Ever wonder how Britain’s most iconic pop pian...</td>\n",
 399 |        "      <td>1</td>\n",
 400 |        "    </tr>\n",
 401 |        "    <tr>\n",
 402 |        "      <th>7</th>\n",
 403 |        "      <td>PARIS  —   France chose an idealistic, traditi...</td>\n",
 404 |        "      <td>0</td>\n",
 405 |        "    </tr>\n",
 406 |        "    <tr>\n",
 407 |        "      <th>8</th>\n",
 408 |        "      <td>Donald J. Trump is scheduled to make a highly ...</td>\n",
 409 |        "      <td>0</td>\n",
 410 |        "    </tr>\n",
 411 |        "    <tr>\n",
 412 |        "      <th>9</th>\n",
 413 |        "      <td>A week before Michael T. Flynn resigned as nat...</td>\n",
 414 |        "      <td>0</td>\n",
 415 |        "    </tr>\n",
 416 |        "    <tr>\n",
 417 |        "      <th>10</th>\n",
 418 |        "      <td>Organizing for Action, the activist group that...</td>\n",
 419 |        "      <td>0</td>\n",
 420 |        "    </tr>\n",
 421 |        "    <tr>\n",
 422 |        "      <th>11</th>\n",
 423 |        "      <td>The BBC produced spoof on the “Real Housewives...</td>\n",
 424 |        "      <td>0</td>\n",
 425 |        "    </tr>\n",
 426 |        "    <tr>\n",
 427 |        "      <th>12</th>\n",
 428 |        "      <td>The mystery surrounding The Third Reich and Na...</td>\n",
 429 |        "      <td>1</td>\n",
 430 |        "    </tr>\n",
 431 |        "    <tr>\n",
 432 |        "      <th>13</th>\n",
 433 |        "      <td>Clinton Campaign Demands FBI Affirm Trump's Ru...</td>\n",
 434 |        "      <td>1</td>\n",
 435 |        "    </tr>\n",
 436 |        "    <tr>\n",
 437 |        "      <th>14</th>\n",
 438 |        "      <td>Yes, There Are Paid Government Trolls On Socia...</td>\n",
 439 |        "      <td>1</td>\n",
 440 |        "    </tr>\n",
 441 |        "  </tbody>\n",
 442 |        "</table>\n",
 443 |        "</div>"
 444 |       ],
 445 |       "text/plain": [
 446 |        "                                                 text  label\n",
 447 |        "0   House Dem Aide: We Didn’t Even See Comey’s Let...      1\n",
 448 |        "1   Ever get the feeling your life circles the rou...      0\n",
 449 |        "2   Why the Truth Might Get You Fired October 29, ...      1\n",
 450 |        "3   Videos 15 Civilians Killed In Single US Airstr...      1\n",
 451 |        "4   Print \\nAn Iranian woman has been sentenced to...      1\n",
 452 |        "5   In these trying times, Jackie Mason is the Voi...      0\n",
 453 |        "6   Ever wonder how Britain’s most iconic pop pian...      1\n",
 454 |        "7   PARIS  —   France chose an idealistic, traditi...      0\n",
 455 |        "8   Donald J. Trump is scheduled to make a highly ...      0\n",
 456 |        "9   A week before Michael T. Flynn resigned as nat...      0\n",
 457 |        "10  Organizing for Action, the activist group that...      0\n",
 458 |        "11  The BBC produced spoof on the “Real Housewives...      0\n",
 459 |        "12  The mystery surrounding The Third Reich and Na...      1\n",
 460 |        "13  Clinton Campaign Demands FBI Affirm Trump's Ru...      1\n",
 461 |        "14  Yes, There Are Paid Government Trolls On Socia...      1"
 462 |       ]
 463 |      },
 464 |      "execution_count": 50,
 465 |      "metadata": {},
 466 |      "output_type": "execute_result"
 467 |     }
 468 |    ],
 469 |    "source": [
 470 |     "train_df.head(15)"
 471 |    ]
 472 |   },
 473 |   {
 474 |    "cell_type": "code",
 475 |    "execution_count": 51,
 476 |    "metadata": {
 477 |     "execution": {
 478 |      "iopub.execute_input": "2021-05-25T06:50:32.144642Z",
 479 |      "iopub.status.busy": "2021-05-25T06:50:32.144299Z",
 480 |      "iopub.status.idle": "2021-05-25T06:50:32.302296Z",
 481 |      "shell.execute_reply": "2021-05-25T06:50:32.301193Z",
 482 |      "shell.execute_reply.started": "2021-05-25T06:50:32.144612Z"
 483 |     }
 484 |    },
 485 |    "outputs": [
 486 |     {
 487 |      "data": {
 488 |       "text/plain": [
 489 |        "<AxesSubplot:xlabel='label', ylabel='count'>"
 490 |       ]
 491 |      },
 492 |      "execution_count": 51,
 493 |      "metadata": {},
 494 |      "output_type": "execute_result"
 495 |     },
 496 |     {
 497 |      "data": {
 498 |       "image/png": "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\n",
 499 |       "text/plain": [
 500 |        "<Figure size 432x288 with 1 Axes>"
 501 |       ]
 502 |      },
 503 |      "metadata": {
 504 |       "needs_background": "light"
 505 |      },
 506 |      "output_type": "display_data"
 507 |     }
 508 |    ],
 509 |    "source": [
 510 |     "def create_distribution(dataFile):\n",
 511 |     "    return sb.countplot(x='label', data=dataFile, palette='hls')\n",
 512 |     "\n",
 513 |     "# by calling below we can see that training, test and valid data seems to be failry evenly distributed between the classes\n",
 514 |     "create_distribution(train_df)"
 515 |    ]
 516 |   },
 517 |   {
 518 |    "cell_type": "code",
 519 |    "execution_count": null,
 520 |    "metadata": {
 521 |     "execution": {
 522 |      "iopub.execute_input": "2021-05-25T06:50:32.306146Z",
 523 |      "iopub.status.busy": "2021-05-25T06:50:32.305826Z",
 524 |      "iopub.status.idle": "2021-05-25T06:50:32.335357Z",
 525 |      "shell.execute_reply": "2021-05-25T06:50:32.33417Z",
 526 |      "shell.execute_reply.started": "2021-05-25T06:50:32.306118Z"
 527 |     }
 528 |    },
 529 |    "outputs": [],
 530 |    "source": [
 531 |     "def data_qualityCheck():\n",
 532 |     "    print(\"Checking data qualitites...\")\n",
 533 |     "    train_df.isnull().sum()\n",
 534 |     "    train_df.info()  \n",
 535 |     "    print(\"check finished.\")\n",
 536 |     "data_qualityCheck()"
 537 |    ]
 538 |   },
 539 |   {
 540 |    "cell_type": "code",
 541 |    "execution_count": 60,
 542 |    "metadata": {
 543 |     "execution": {
 544 |      "iopub.execute_input": "2021-05-25T06:50:32.337061Z",
 545 |      "iopub.status.busy": "2021-05-25T06:50:32.336735Z",
 546 |      "iopub.status.idle": "2021-05-25T06:50:32.367948Z",
 547 |      "shell.execute_reply": "2021-05-25T06:50:32.366933Z",
 548 |      "shell.execute_reply.started": "2021-05-25T06:50:32.33703Z"
 549 |     }
 550 |    },
 551 |    "outputs": [],
 552 |    "source": [
 553 |     "train_df = train_df.dropna()"
 554 |    ]
 555 |   },
 556 |   {
 557 |    "cell_type": "code",
 558 |    "execution_count": 61,
 559 |    "metadata": {
 560 |     "execution": {
 561 |      "iopub.execute_input": "2021-05-25T06:50:32.369967Z",
 562 |      "iopub.status.busy": "2021-05-25T06:50:32.369528Z",
 563 |      "iopub.status.idle": "2021-05-25T06:50:32.399296Z",
 564 |      "shell.execute_reply": "2021-05-25T06:50:32.398264Z",
 565 |      "shell.execute_reply.started": "2021-05-25T06:50:32.369924Z"
 566 |     }
 567 |    },
 568 |    "outputs": [
 569 |     {
 570 |      "name": "stdout",
 571 |      "output_type": "stream",
 572 |      "text": [
 573 |       "Checking data qualitites...\n",
 574 |       "<class 'pandas.core.frame.DataFrame'>\n",
 575 |       "Int64Index: 20761 entries, 0 to 20799\n",
 576 |       "Data columns (total 2 columns):\n",
 577 |       " #   Column  Non-Null Count  Dtype \n",
 578 |       "---  ------  --------------  ----- \n",
 579 |       " 0   text    20761 non-null  object\n",
 580 |       " 1   label   20761 non-null  int64 \n",
 581 |       "dtypes: int64(1), object(1)\n",
 582 |       "memory usage: 486.6+ KB\n",
 583 |       "check finished.\n"
 584 |      ]
 585 |     }
 586 |    ],
 587 |    "source": [
 588 |     "data_qualityCheck()"
 589 |    ]
 590 |   },
 591 |   {
 592 |    "cell_type": "code",
 593 |    "execution_count": 62,
 594 |    "metadata": {
 595 |     "execution": {
 596 |      "iopub.execute_input": "2021-05-25T06:50:32.401314Z",
 597 |      "iopub.status.busy": "2021-05-25T06:50:32.400868Z",
 598 |      "iopub.status.idle": "2021-05-25T06:50:32.407806Z",
 599 |      "shell.execute_reply": "2021-05-25T06:50:32.406589Z",
 600 |      "shell.execute_reply.started": "2021-05-25T06:50:32.401272Z"
 601 |     }
 602 |    },
 603 |    "outputs": [
 604 |     {
 605 |      "data": {
 606 |       "text/plain": [
 607 |        "(20761, 2)"
 608 |       ]
 609 |      },
 610 |      "execution_count": 62,
 611 |      "metadata": {},
 612 |      "output_type": "execute_result"
 613 |     }
 614 |    ],
 615 |    "source": [
 616 |     "train_df.shape"
 617 |    ]
 618 |   },
 619 |   {
 620 |    "cell_type": "code",
 621 |    "execution_count": 63,
 622 |    "metadata": {
 623 |     "execution": {
 624 |      "iopub.execute_input": "2021-05-25T06:50:32.409912Z",
 625 |      "iopub.status.busy": "2021-05-25T06:50:32.409162Z",
 626 |      "iopub.status.idle": "2021-05-25T06:50:32.426843Z",
 627 |      "shell.execute_reply": "2021-05-25T06:50:32.425727Z",
 628 |      "shell.execute_reply.started": "2021-05-25T06:50:32.409868Z"
 629 |     }
 630 |    },
 631 |    "outputs": [
 632 |     {
 633 |      "data": {
 634 |       "text/html": [
 635 |        "<div>\n",
 636 |        "<style scoped>\n",
 637 |        "    .dataframe tbody tr th:only-of-type {\n",
 638 |        "        vertical-align: middle;\n",
 639 |        "    }\n",
 640 |        "\n",
 641 |        "    .dataframe tbody tr th {\n",
 642 |        "        vertical-align: top;\n",
 643 |        "    }\n",
 644 |        "\n",
 645 |        "    .dataframe thead th {\n",
 646 |        "        text-align: right;\n",
 647 |        "    }\n",
 648 |        "</style>\n",
 649 |        "<table border=\"1\" class=\"dataframe\">\n",
 650 |        "  <thead>\n",
 651 |        "    <tr style=\"text-align: right;\">\n",
 652 |        "      <th></th>\n",
 653 |        "      <th>text</th>\n",
 654 |        "      <th>label</th>\n",
 655 |        "    </tr>\n",
 656 |        "  </thead>\n",
 657 |        "  <tbody>\n",
 658 |        "    <tr>\n",
 659 |        "      <th>0</th>\n",
 660 |        "      <td>House Dem Aide: We Didn’t Even See Comey’s Let...</td>\n",
 661 |        "      <td>1</td>\n",
 662 |        "    </tr>\n",
 663 |        "    <tr>\n",
 664 |        "      <th>1</th>\n",
 665 |        "      <td>Ever get the feeling your life circles the rou...</td>\n",
 666 |        "      <td>0</td>\n",
 667 |        "    </tr>\n",
 668 |        "    <tr>\n",
 669 |        "      <th>2</th>\n",
 670 |        "      <td>Why the Truth Might Get You Fired October 29, ...</td>\n",
 671 |        "      <td>1</td>\n",
 672 |        "    </tr>\n",
 673 |        "    <tr>\n",
 674 |        "      <th>3</th>\n",
 675 |        "      <td>Videos 15 Civilians Killed In Single US Airstr...</td>\n",
 676 |        "      <td>1</td>\n",
 677 |        "    </tr>\n",
 678 |        "    <tr>\n",
 679 |        "      <th>4</th>\n",
 680 |        "      <td>Print \\nAn Iranian woman has been sentenced to...</td>\n",
 681 |        "      <td>1</td>\n",
 682 |        "    </tr>\n",
 683 |        "    <tr>\n",
 684 |        "      <th>5</th>\n",
 685 |        "      <td>In these trying times, Jackie Mason is the Voi...</td>\n",
 686 |        "      <td>0</td>\n",
 687 |        "    </tr>\n",
 688 |        "    <tr>\n",
 689 |        "      <th>6</th>\n",
 690 |        "      <td>Ever wonder how Britain’s most iconic pop pian...</td>\n",
 691 |        "      <td>1</td>\n",
 692 |        "    </tr>\n",
 693 |        "    <tr>\n",
 694 |        "      <th>7</th>\n",
 695 |        "      <td>PARIS  —   France chose an idealistic, traditi...</td>\n",
 696 |        "      <td>0</td>\n",
 697 |        "    </tr>\n",
 698 |        "    <tr>\n",
 699 |        "      <th>8</th>\n",
 700 |        "      <td>Donald J. Trump is scheduled to make a highly ...</td>\n",
 701 |        "      <td>0</td>\n",
 702 |        "    </tr>\n",
 703 |        "    <tr>\n",
 704 |        "      <th>9</th>\n",
 705 |        "      <td>A week before Michael T. Flynn resigned as nat...</td>\n",
 706 |        "      <td>0</td>\n",
 707 |        "    </tr>\n",
 708 |        "  </tbody>\n",
 709 |        "</table>\n",
 710 |        "</div>"
 711 |       ],
 712 |       "text/plain": [
 713 |        "                                                text  label\n",
 714 |        "0  House Dem Aide: We Didn’t Even See Comey’s Let...      1\n",
 715 |        "1  Ever get the feeling your life circles the rou...      0\n",
 716 |        "2  Why the Truth Might Get You Fired October 29, ...      1\n",
 717 |        "3  Videos 15 Civilians Killed In Single US Airstr...      1\n",
 718 |        "4  Print \\nAn Iranian woman has been sentenced to...      1\n",
 719 |        "5  In these trying times, Jackie Mason is the Voi...      0\n",
 720 |        "6  Ever wonder how Britain’s most iconic pop pian...      1\n",
 721 |        "7  PARIS  —   France chose an idealistic, traditi...      0\n",
 722 |        "8  Donald J. Trump is scheduled to make a highly ...      0\n",
 723 |        "9  A week before Michael T. Flynn resigned as nat...      0"
 724 |       ]
 725 |      },
 726 |      "execution_count": 63,
 727 |      "metadata": {},
 728 |      "output_type": "execute_result"
 729 |     }
 730 |    ],
 731 |    "source": [
 732 |     "train_df.head(10)"
 733 |    ]
 734 |   },
 735 |   {
 736 |    "cell_type": "code",
 737 |    "execution_count": 64,
 738 |    "metadata": {
 739 |     "execution": {
 740 |      "iopub.execute_input": "2021-05-25T06:50:32.42892Z",
 741 |      "iopub.status.busy": "2021-05-25T06:50:32.42853Z",
 742 |      "iopub.status.idle": "2021-05-25T06:50:32.433555Z",
 743 |      "shell.execute_reply": "2021-05-25T06:50:32.432625Z",
 744 |      "shell.execute_reply.started": "2021-05-25T06:50:32.428889Z"
 745 |     }
 746 |    },
 747 |    "outputs": [],
 748 |    "source": [
 749 |     "train_df.reset_index(drop= True,inplace=True)"
 750 |    ]
 751 |   },
 752 |   {
 753 |    "cell_type": "code",
 754 |    "execution_count": 65,
 755 |    "metadata": {
 756 |     "execution": {
 757 |      "iopub.execute_input": "2021-05-25T06:50:32.435265Z",
 758 |      "iopub.status.busy": "2021-05-25T06:50:32.434781Z",
 759 |      "iopub.status.idle": "2021-05-25T06:50:32.454591Z",
 760 |      "shell.execute_reply": "2021-05-25T06:50:32.453613Z",
 761 |      "shell.execute_reply.started": "2021-05-25T06:50:32.435224Z"
 762 |     }
 763 |    },
 764 |    "outputs": [
 765 |     {
 766 |      "data": {
 767 |       "text/html": [
 768 |        "<div>\n",
 769 |        "<style scoped>\n",
 770 |        "    .dataframe tbody tr th:only-of-type {\n",
 771 |        "        vertical-align: middle;\n",
 772 |        "    }\n",
 773 |        "\n",
 774 |        "    .dataframe tbody tr th {\n",
 775 |        "        vertical-align: top;\n",
 776 |        "    }\n",
 777 |        "\n",
 778 |        "    .dataframe thead th {\n",
 779 |        "        text-align: right;\n",
 780 |        "    }\n",
 781 |        "</style>\n",
 782 |        "<table border=\"1\" class=\"dataframe\">\n",
 783 |        "  <thead>\n",
 784 |        "    <tr style=\"text-align: right;\">\n",
 785 |        "      <th></th>\n",
 786 |        "      <th>text</th>\n",
 787 |        "      <th>label</th>\n",
 788 |        "    </tr>\n",
 789 |        "  </thead>\n",
 790 |        "  <tbody>\n",
 791 |        "    <tr>\n",
 792 |        "      <th>0</th>\n",
 793 |        "      <td>House Dem Aide: We Didn’t Even See Comey’s Let...</td>\n",
 794 |        "      <td>1</td>\n",
 795 |        "    </tr>\n",
 796 |        "    <tr>\n",
 797 |        "      <th>1</th>\n",
 798 |        "      <td>Ever get the feeling your life circles the rou...</td>\n",
 799 |        "      <td>0</td>\n",
 800 |        "    </tr>\n",
 801 |        "    <tr>\n",
 802 |        "      <th>2</th>\n",
 803 |        "      <td>Why the Truth Might Get You Fired October 29, ...</td>\n",
 804 |        "      <td>1</td>\n",
 805 |        "    </tr>\n",
 806 |        "    <tr>\n",
 807 |        "      <th>3</th>\n",
 808 |        "      <td>Videos 15 Civilians Killed In Single US Airstr...</td>\n",
 809 |        "      <td>1</td>\n",
 810 |        "    </tr>\n",
 811 |        "    <tr>\n",
 812 |        "      <th>4</th>\n",
 813 |        "      <td>Print \\nAn Iranian woman has been sentenced to...</td>\n",
 814 |        "      <td>1</td>\n",
 815 |        "    </tr>\n",
 816 |        "    <tr>\n",
 817 |        "      <th>5</th>\n",
 818 |        "      <td>In these trying times, Jackie Mason is the Voi...</td>\n",
 819 |        "      <td>0</td>\n",
 820 |        "    </tr>\n",
 821 |        "    <tr>\n",
 822 |        "      <th>6</th>\n",
 823 |        "      <td>Ever wonder how Britain’s most iconic pop pian...</td>\n",
 824 |        "      <td>1</td>\n",
 825 |        "    </tr>\n",
 826 |        "    <tr>\n",
 827 |        "      <th>7</th>\n",
 828 |        "      <td>PARIS  —   France chose an idealistic, traditi...</td>\n",
 829 |        "      <td>0</td>\n",
 830 |        "    </tr>\n",
 831 |        "    <tr>\n",
 832 |        "      <th>8</th>\n",
 833 |        "      <td>Donald J. Trump is scheduled to make a highly ...</td>\n",
 834 |        "      <td>0</td>\n",
 835 |        "    </tr>\n",
 836 |        "    <tr>\n",
 837 |        "      <th>9</th>\n",
 838 |        "      <td>A week before Michael T. Flynn resigned as nat...</td>\n",
 839 |        "      <td>0</td>\n",
 840 |        "    </tr>\n",
 841 |        "  </tbody>\n",
 842 |        "</table>\n",
 843 |        "</div>"
 844 |       ],
 845 |       "text/plain": [
 846 |        "                                                text  label\n",
 847 |        "0  House Dem Aide: We Didn’t Even See Comey’s Let...      1\n",
 848 |        "1  Ever get the feeling your life circles the rou...      0\n",
 849 |        "2  Why the Truth Might Get You Fired October 29, ...      1\n",
 850 |        "3  Videos 15 Civilians Killed In Single US Airstr...      1\n",
 851 |        "4  Print \\nAn Iranian woman has been sentenced to...      1\n",
 852 |        "5  In these trying times, Jackie Mason is the Voi...      0\n",
 853 |        "6  Ever wonder how Britain’s most iconic pop pian...      1\n",
 854 |        "7  PARIS  —   France chose an idealistic, traditi...      0\n",
 855 |        "8  Donald J. Trump is scheduled to make a highly ...      0\n",
 856 |        "9  A week before Michael T. Flynn resigned as nat...      0"
 857 |       ]
 858 |      },
 859 |      "execution_count": 65,
 860 |      "metadata": {},
 861 |      "output_type": "execute_result"
 862 |     }
 863 |    ],
 864 |    "source": [
 865 |     "train_df.head(10)"
 866 |    ]
 867 |   },
 868 |   {
 869 |    "cell_type": "code",
 870 |    "execution_count": 67,
 871 |    "metadata": {
 872 |     "execution": {
 873 |      "iopub.execute_input": "2021-05-25T06:50:32.457112Z",
 874 |      "iopub.status.busy": "2021-05-25T06:50:32.45653Z",
 875 |      "iopub.status.idle": "2021-05-25T06:50:32.46346Z",
 876 |      "shell.execute_reply": "2021-05-25T06:50:32.461467Z",
 877 |      "shell.execute_reply.started": "2021-05-25T06:50:32.457067Z"
 878 |     }
 879 |    },
 880 |    "outputs": [],
 881 |    "source": [
 882 |     "label_train = train_df.label"
 883 |    ]
 884 |   },
 885 |   {
 886 |    "cell_type": "code",
 887 |    "execution_count": 68,
 888 |    "metadata": {
 889 |     "execution": {
 890 |      "iopub.execute_input": "2021-05-25T06:50:32.46513Z",
 891 |      "iopub.status.busy": "2021-05-25T06:50:32.46484Z",
 892 |      "iopub.status.idle": "2021-05-25T06:50:32.479833Z",
 893 |      "shell.execute_reply": "2021-05-25T06:50:32.478601Z",
 894 |      "shell.execute_reply.started": "2021-05-25T06:50:32.465102Z"
 895 |     }
 896 |    },
 897 |    "outputs": [
 898 |     {
 899 |      "data": {
 900 |       "text/plain": [
 901 |        "0    1\n",
 902 |        "1    0\n",
 903 |        "2    1\n",
 904 |        "3    1\n",
 905 |        "4    1\n",
 906 |        "5    0\n",
 907 |        "6    1\n",
 908 |        "7    0\n",
 909 |        "8    0\n",
 910 |        "9    0\n",
 911 |        "Name: label, dtype: int64"
 912 |       ]
 913 |      },
 914 |      "execution_count": 68,
 915 |      "metadata": {},
 916 |      "output_type": "execute_result"
 917 |     }
 918 |    ],
 919 |    "source": [
 920 |     "label_train.head(10)"
 921 |    ]
 922 |   },
 923 |   {
 924 |    "cell_type": "code",
 925 |    "execution_count": 69,
 926 |    "metadata": {
 927 |     "execution": {
 928 |      "iopub.execute_input": "2021-05-25T06:50:32.481757Z",
 929 |      "iopub.status.busy": "2021-05-25T06:50:32.481439Z",
 930 |      "iopub.status.idle": "2021-05-25T06:50:32.493571Z",
 931 |      "shell.execute_reply": "2021-05-25T06:50:32.492736Z",
 932 |      "shell.execute_reply.started": "2021-05-25T06:50:32.481728Z"
 933 |     }
 934 |    },
 935 |    "outputs": [],
 936 |    "source": [
 937 |     "train_df = train_df.drop(\"label\", axis = 1)"
 938 |    ]
 939 |   },
 940 |   {
 941 |    "cell_type": "code",
 942 |    "execution_count": 70,
 943 |    "metadata": {
 944 |     "execution": {
 945 |      "iopub.execute_input": "2021-05-25T06:50:32.495566Z",
 946 |      "iopub.status.busy": "2021-05-25T06:50:32.495116Z",
 947 |      "iopub.status.idle": "2021-05-25T06:50:32.513957Z",
 948 |      "shell.execute_reply": "2021-05-25T06:50:32.51265Z",
 949 |      "shell.execute_reply.started": "2021-05-25T06:50:32.495526Z"
 950 |     }
 951 |    },
 952 |    "outputs": [
 953 |     {
 954 |      "data": {
 955 |       "text/html": [
 956 |        "<div>\n",
 957 |        "<style scoped>\n",
 958 |        "    .dataframe tbody tr th:only-of-type {\n",
 959 |        "        vertical-align: middle;\n",
 960 |        "    }\n",
 961 |        "\n",
 962 |        "    .dataframe tbody tr th {\n",
 963 |        "        vertical-align: top;\n",
 964 |        "    }\n",
 965 |        "\n",
 966 |        "    .dataframe thead th {\n",
 967 |        "        text-align: right;\n",
 968 |        "    }\n",
 969 |        "</style>\n",
 970 |        "<table border=\"1\" class=\"dataframe\">\n",
 971 |        "  <thead>\n",
 972 |        "    <tr style=\"text-align: right;\">\n",
 973 |        "      <th></th>\n",
 974 |        "      <th>text</th>\n",
 975 |        "    </tr>\n",
 976 |        "  </thead>\n",
 977 |        "  <tbody>\n",
 978 |        "    <tr>\n",
 979 |        "      <th>0</th>\n",
 980 |        "      <td>House Dem Aide: We Didn’t Even See Comey’s Let...</td>\n",
 981 |        "    </tr>\n",
 982 |        "    <tr>\n",
 983 |        "      <th>1</th>\n",
 984 |        "      <td>Ever get the feeling your life circles the rou...</td>\n",
 985 |        "    </tr>\n",
 986 |        "    <tr>\n",
 987 |        "      <th>2</th>\n",
 988 |        "      <td>Why the Truth Might Get You Fired October 29, ...</td>\n",
 989 |        "    </tr>\n",
 990 |        "    <tr>\n",
 991 |        "      <th>3</th>\n",
 992 |        "      <td>Videos 15 Civilians Killed In Single US Airstr...</td>\n",
 993 |        "    </tr>\n",
 994 |        "    <tr>\n",
 995 |        "      <th>4</th>\n",
 996 |        "      <td>Print \\nAn Iranian woman has been sentenced to...</td>\n",
 997 |        "    </tr>\n",
 998 |        "    <tr>\n",
 999 |        "      <th>5</th>\n",
1000 |        "      <td>In these trying times, Jackie Mason is the Voi...</td>\n",
1001 |        "    </tr>\n",
1002 |        "    <tr>\n",
1003 |        "      <th>6</th>\n",
1004 |        "      <td>Ever wonder how Britain’s most iconic pop pian...</td>\n",
1005 |        "    </tr>\n",
1006 |        "    <tr>\n",
1007 |        "      <th>7</th>\n",
1008 |        "      <td>PARIS  —   France chose an idealistic, traditi...</td>\n",
1009 |        "    </tr>\n",
1010 |        "    <tr>\n",
1011 |        "      <th>8</th>\n",
1012 |        "      <td>Donald J. Trump is scheduled to make a highly ...</td>\n",
1013 |        "    </tr>\n",
1014 |        "    <tr>\n",
1015 |        "      <th>9</th>\n",
1016 |        "      <td>A week before Michael T. Flynn resigned as nat...</td>\n",
1017 |        "    </tr>\n",
1018 |        "  </tbody>\n",
1019 |        "</table>\n",
1020 |        "</div>"
1021 |       ],
1022 |       "text/plain": [
1023 |        "                                                text\n",
1024 |        "0  House Dem Aide: We Didn’t Even See Comey’s Let...\n",
1025 |        "1  Ever get the feeling your life circles the rou...\n",
1026 |        "2  Why the Truth Might Get You Fired October 29, ...\n",
1027 |        "3  Videos 15 Civilians Killed In Single US Airstr...\n",
1028 |        "4  Print \\nAn Iranian woman has been sentenced to...\n",
1029 |        "5  In these trying times, Jackie Mason is the Voi...\n",
1030 |        "6  Ever wonder how Britain’s most iconic pop pian...\n",
1031 |        "7  PARIS  —   France chose an idealistic, traditi...\n",
1032 |        "8  Donald J. Trump is scheduled to make a highly ...\n",
1033 |        "9  A week before Michael T. Flynn resigned as nat..."
1034 |       ]
1035 |      },
1036 |      "execution_count": 70,
1037 |      "metadata": {},
1038 |      "output_type": "execute_result"
1039 |     }
1040 |    ],
1041 |    "source": [
1042 |     "train_df.head(10)"
1043 |    ]
1044 |   },
1045 |   {
1046 |    "cell_type": "code",
1047 |    "execution_count": 59,
1048 |    "metadata": {},
1049 |    "outputs": [
1050 |     {
1051 |      "data": {
1052 |       "text/plain": [
1053 |        "'Comments Actor Steven Seagal has just stood up for America, while the rest of Hollywood remains silent. This week has been rough for our country. First, Democratic nominee Hillary Clinton collapsed at the 9/11 memorial. Then, she called millions of hardworking Americans “deplorable.” As if that weren’t enough, NFL players throughout the country have been blatantly disrespecting the American flag. Needless to say, Seagal had enough. “I think the most important job as Secretary of State is ensuring our people don’t get killed,” Seagal tweeted. “If you can’t do [email protected] “Pneumonia or not, she is going to be disastrous for the American people. #NoToHillary,” he continued. Of course, Seagal quickly became the target of liberal fire for his comments, but he refused to break down. He particularly lost it when one Twitter user tried to argued that Hillary was the most capable for the presidency. “Capable? Capable of leaving AMERICANS to die. Capable of disregarding law. Capable of disrespecting rape survivors,” he argued. He then went on to address race relations in the United States, and the true role President Barack Obama has played in the social evolution of this country. “Obama has been abysmal for race relations in USA. #TRUTH We need to start having honest dialog,” he wrote. Seagal concluded by pointing out the irony of the attacks he was receiving from liberals everywhere. “Best thing in the world…Making one statement about freedom and getting attacked by every Demo out there… #HYPOCRITICAL,” he tweeted. “An America without a democrat in the White House, is a SAFER America.” What do you think of Seagal’s comments? '"
1054 |       ]
1055 |      },
1056 |      "execution_count": 59,
1057 |      "metadata": {},
1058 |      "output_type": "execute_result"
1059 |     }
1060 |    ],
1061 |    "source": [
1062 |     "train_df['text'][2188]"
1063 |    ]
1064 |   },
1065 |   {
1066 |    "cell_type": "code",
1067 |    "execution_count": 71,
1068 |    "metadata": {
1069 |     "execution": {
1070 |      "iopub.execute_input": "2021-05-25T06:50:32.51602Z",
1071 |      "iopub.status.busy": "2021-05-25T06:50:32.515411Z",
1072 |      "iopub.status.idle": "2021-05-25T06:50:32.531829Z",
1073 |      "shell.execute_reply": "2021-05-25T06:50:32.530895Z",
1074 |      "shell.execute_reply.started": "2021-05-25T06:50:32.515972Z"
1075 |     }
1076 |    },
1077 |    "outputs": [],
1078 |    "source": [
1079 |     "lemmatizer = WordNetLemmatizer()\n",
1080 |     "stpwrds = list(stopwords.words('english'))"
1081 |    ]
1082 |   },
1083 |   {
1084 |    "cell_type": "code",
1085 |    "execution_count": 73,
1086 |    "metadata": {},
1087 |    "outputs": [
1088 |     {
1089 |      "data": {
1090 |       "text/plain": [
1091 |        "['i',\n",
1092 |        " 'me',\n",
1093 |        " 'my',\n",
1094 |        " 'myself',\n",
1095 |        " 'we',\n",
1096 |        " 'our',\n",
1097 |        " 'ours',\n",
1098 |        " 'ourselves',\n",
1099 |        " 'you',\n",
1100 |        " \"you're\",\n",
1101 |        " \"you've\",\n",
1102 |        " \"you'll\",\n",
1103 |        " \"you'd\",\n",
1104 |        " 'your',\n",
1105 |        " 'yours',\n",
1106 |        " 'yourself',\n",
1107 |        " 'yourselves',\n",
1108 |        " 'he',\n",
1109 |        " 'him',\n",
1110 |        " 'his',\n",
1111 |        " 'himself',\n",
1112 |        " 'she',\n",
1113 |        " \"she's\",\n",
1114 |        " 'her',\n",
1115 |        " 'hers',\n",
1116 |        " 'herself',\n",
1117 |        " 'it',\n",
1118 |        " \"it's\",\n",
1119 |        " 'its',\n",
1120 |        " 'itself',\n",
1121 |        " 'they',\n",
1122 |        " 'them',\n",
1123 |        " 'their',\n",
1124 |        " 'theirs',\n",
1125 |        " 'themselves',\n",
1126 |        " 'what',\n",
1127 |        " 'which',\n",
1128 |        " 'who',\n",
1129 |        " 'whom',\n",
1130 |        " 'this',\n",
1131 |        " 'that',\n",
1132 |        " \"that'll\",\n",
1133 |        " 'these',\n",
1134 |        " 'those',\n",
1135 |        " 'am',\n",
1136 |        " 'is',\n",
1137 |        " 'are',\n",
1138 |        " 'was',\n",
1139 |        " 'were',\n",
1140 |        " 'be',\n",
1141 |        " 'been',\n",
1142 |        " 'being',\n",
1143 |        " 'have',\n",
1144 |        " 'has',\n",
1145 |        " 'had',\n",
1146 |        " 'having',\n",
1147 |        " 'do',\n",
1148 |        " 'does',\n",
1149 |        " 'did',\n",
1150 |        " 'doing',\n",
1151 |        " 'a',\n",
1152 |        " 'an',\n",
1153 |        " 'the',\n",
1154 |        " 'and',\n",
1155 |        " 'but',\n",
1156 |        " 'if',\n",
1157 |        " 'or',\n",
1158 |        " 'because',\n",
1159 |        " 'as',\n",
1160 |        " 'until',\n",
1161 |        " 'while',\n",
1162 |        " 'of',\n",
1163 |        " 'at',\n",
1164 |        " 'by',\n",
1165 |        " 'for',\n",
1166 |        " 'with',\n",
1167 |        " 'about',\n",
1168 |        " 'against',\n",
1169 |        " 'between',\n",
1170 |        " 'into',\n",
1171 |        " 'through',\n",
1172 |        " 'during',\n",
1173 |        " 'before',\n",
1174 |        " 'after',\n",
1175 |        " 'above',\n",
1176 |        " 'below',\n",
1177 |        " 'to',\n",
1178 |        " 'from',\n",
1179 |        " 'up',\n",
1180 |        " 'down',\n",
1181 |        " 'in',\n",
1182 |        " 'out',\n",
1183 |        " 'on',\n",
1184 |        " 'off',\n",
1185 |        " 'over',\n",
1186 |        " 'under',\n",
1187 |        " 'again',\n",
1188 |        " 'further',\n",
1189 |        " 'then',\n",
1190 |        " 'once',\n",
1191 |        " 'here',\n",
1192 |        " 'there',\n",
1193 |        " 'when',\n",
1194 |        " 'where',\n",
1195 |        " 'why',\n",
1196 |        " 'how',\n",
1197 |        " 'all',\n",
1198 |        " 'any',\n",
1199 |        " 'both',\n",
1200 |        " 'each',\n",
1201 |        " 'few',\n",
1202 |        " 'more',\n",
1203 |        " 'most',\n",
1204 |        " 'other',\n",
1205 |        " 'some',\n",
1206 |        " 'such',\n",
1207 |        " 'no',\n",
1208 |        " 'nor',\n",
1209 |        " 'not',\n",
1210 |        " 'only',\n",
1211 |        " 'own',\n",
1212 |        " 'same',\n",
1213 |        " 'so',\n",
1214 |        " 'than',\n",
1215 |        " 'too',\n",
1216 |        " 'very',\n",
1217 |        " 's',\n",
1218 |        " 't',\n",
1219 |        " 'can',\n",
1220 |        " 'will',\n",
1221 |        " 'just',\n",
1222 |        " 'don',\n",
1223 |        " \"don't\",\n",
1224 |        " 'should',\n",
1225 |        " \"should've\",\n",
1226 |        " 'now',\n",
1227 |        " 'd',\n",
1228 |        " 'll',\n",
1229 |        " 'm',\n",
1230 |        " 'o',\n",
1231 |        " 're',\n",
1232 |        " 've',\n",
1233 |        " 'y',\n",
1234 |        " 'ain',\n",
1235 |        " 'aren',\n",
1236 |        " \"aren't\",\n",
1237 |        " 'couldn',\n",
1238 |        " \"couldn't\",\n",
1239 |        " 'didn',\n",
1240 |        " \"didn't\",\n",
1241 |        " 'doesn',\n",
1242 |        " \"doesn't\",\n",
1243 |        " 'hadn',\n",
1244 |        " \"hadn't\",\n",
1245 |        " 'hasn',\n",
1246 |        " \"hasn't\",\n",
1247 |        " 'haven',\n",
1248 |        " \"haven't\",\n",
1249 |        " 'isn',\n",
1250 |        " \"isn't\",\n",
1251 |        " 'ma',\n",
1252 |        " 'mightn',\n",
1253 |        " \"mightn't\",\n",
1254 |        " 'mustn',\n",
1255 |        " \"mustn't\",\n",
1256 |        " 'needn',\n",
1257 |        " \"needn't\",\n",
1258 |        " 'shan',\n",
1259 |        " \"shan't\",\n",
1260 |        " 'shouldn',\n",
1261 |        " \"shouldn't\",\n",
1262 |        " 'wasn',\n",
1263 |        " \"wasn't\",\n",
1264 |        " 'weren',\n",
1265 |        " \"weren't\",\n",
1266 |        " 'won',\n",
1267 |        " \"won't\",\n",
1268 |        " 'wouldn',\n",
1269 |        " \"wouldn't\"]"
1270 |       ]
1271 |      },
1272 |      "execution_count": 73,
1273 |      "metadata": {},
1274 |      "output_type": "execute_result"
1275 |     }
1276 |    ],
1277 |    "source": [
1278 |     "stpwrds"
1279 |    ]
1280 |   },
1281 |   {
1282 |    "cell_type": "code",
1283 |    "execution_count": 75,
1284 |    "metadata": {
1285 |     "execution": {
1286 |      "iopub.execute_input": "2021-05-25T06:50:32.54905Z",
1287 |      "iopub.status.busy": "2021-05-25T06:50:32.548517Z",
1288 |      "iopub.status.idle": "2021-05-25T06:53:51.648153Z",
1289 |      "shell.execute_reply": "2021-05-25T06:53:51.647283Z",
1290 |      "shell.execute_reply.started": "2021-05-25T06:50:32.549015Z"
1291 |     }
1292 |    },
1293 |    "outputs": [],
1294 |    "source": [
1295 |     "for x in range(len(train_df)) :\n",
1296 |     "    corpus = []\n",
1297 |     "    review = train_df['text'][x]\n",
1298 |     "    review = re.sub(r'[^a-zA-Z\\s]', '', review)\n",
1299 |     "    review = review.lower()\n",
1300 |     "    review = nltk.word_tokenize(review)\n",
1301 |     "    for y in review :\n",
1302 |     "        if y not in stpwrds :\n",
1303 |     "            corpus.append(lemmatizer.lemmatize(y))\n",
1304 |     "    review = ' '.join(corpus)\n",
1305 |     "    train_df['text'][x] = review      "
1306 |    ]
1307 |   },
1308 |   {
1309 |    "cell_type": "code",
1310 |    "execution_count": 79,
1311 |    "metadata": {
1312 |     "execution": {
1313 |      "iopub.execute_input": "2021-05-25T07:14:51.798724Z",
1314 |      "iopub.status.busy": "2021-05-25T07:14:51.798361Z",
1315 |      "iopub.status.idle": "2021-05-25T07:14:51.805617Z",
1316 |      "shell.execute_reply": "2021-05-25T07:14:51.804946Z",
1317 |      "shell.execute_reply.started": "2021-05-25T07:14:51.798694Z"
1318 |     },
1319 |     "scrolled": true
1320 |    },
1321 |    "outputs": [
1322 |     {
1323 |      "data": {
1324 |       "text/plain": [
1325 |        "'comment actor steven seagal stood america rest hollywood remains silent week rough country first democratic nominee hillary clinton collapsed memorial called million hardworking american deplorable werent enough nfl player throughout country blatantly disrespecting american flag needle say seagal enough think important job secretary state ensuring people dont get killed seagal tweeted cant email protected pneumonia going disastrous american people notohillary continued course seagal quickly became target liberal fire comment refused break particularly lost one twitter user tried argued hillary capable presidency capable capable leaving american die capable disregarding law capable disrespecting rape survivor argued went address race relation united state true role president barack obama played social evolution country obama abysmal race relation usa truth need start honest dialog wrote seagal concluded pointing irony attack receiving liberal everywhere best thing worldmaking one statement freedom getting attacked every demo hypocritical tweeted america without democrat white house safer america think seagals comment'"
1326 |       ]
1327 |      },
1328 |      "execution_count": 79,
1329 |      "metadata": {},
1330 |      "output_type": "execute_result"
1331 |     }
1332 |    ],
1333 |    "source": [
1334 |     "train_df['text'][2182]"
1335 |    ]
1336 |   },
1337 |   {
1338 |    "cell_type": "code",
1339 |    "execution_count": 80,
1340 |    "metadata": {
1341 |     "execution": {
1342 |      "iopub.execute_input": "2021-05-25T07:16:37.152728Z",
1343 |      "iopub.status.busy": "2021-05-25T07:16:37.152216Z",
1344 |      "iopub.status.idle": "2021-05-25T07:16:37.163059Z",
1345 |      "shell.execute_reply": "2021-05-25T07:16:37.161884Z",
1346 |      "shell.execute_reply.started": "2021-05-25T07:16:37.152696Z"
1347 |     }
1348 |    },
1349 |    "outputs": [],
1350 |    "source": [
1351 |     "X_train, X_test, Y_train, Y_test = train_test_split(train_df['text'], label_train, test_size=0.3, random_state=1)"
1352 |    ]
1353 |   },
1354 |   {
1355 |    "cell_type": "code",
1356 |    "execution_count": 81,
1357 |    "metadata": {
1358 |     "scrolled": true
1359 |    },
1360 |    "outputs": [
1361 |     {
1362 |      "data": {
1363 |       "text/plain": [
1364 |        "10140    comment fox news star megyn kelly finally reve...\n",
1365 |        "11508    written daniel mcadams told attack iraq saddam...\n",
1366 |        "4035     u representative california introduced bill bl...\n",
1367 |        "4528     paris salah abdeslam thought direct participan...\n",
1368 |        "2608     photo jorge lascar photo great wall china vasi...\n",
1369 |        "                               ...                        \n",
1370 |        "10955    daily caller progress unity found progress uni...\n",
1371 |        "17289    archie elam third career transition living sta...\n",
1372 |        "5192     philadelphia sometimes exhibition planned year...\n",
1373 |        "12172    here something interesting unz review recipien...\n",
1374 |        "235      changing montenegrin leader change ideology so...\n",
1375 |        "Name: text, Length: 14532, dtype: object"
1376 |       ]
1377 |      },
1378 |      "execution_count": 81,
1379 |      "metadata": {},
1380 |      "output_type": "execute_result"
1381 |     }
1382 |    ],
1383 |    "source": [
1384 |     "X_train"
1385 |    ]
1386 |   },
1387 |   {
1388 |    "cell_type": "code",
1389 |    "execution_count": 83,
1390 |    "metadata": {
1391 |     "execution": {
1392 |      "iopub.execute_input": "2021-05-25T07:17:50.592597Z",
1393 |      "iopub.status.busy": "2021-05-25T07:17:50.592095Z",
1394 |      "iopub.status.idle": "2021-05-25T07:17:50.598862Z",
1395 |      "shell.execute_reply": "2021-05-25T07:17:50.597641Z",
1396 |      "shell.execute_reply.started": "2021-05-25T07:17:50.592566Z"
1397 |     }
1398 |    },
1399 |    "outputs": [
1400 |     {
1401 |      "data": {
1402 |       "text/plain": [
1403 |        "(14532,)"
1404 |       ]
1405 |      },
1406 |      "execution_count": 83,
1407 |      "metadata": {},
1408 |      "output_type": "execute_result"
1409 |     }
1410 |    ],
1411 |    "source": [
1412 |     "X_train.shape"
1413 |    ]
1414 |   },
1415 |   {
1416 |    "cell_type": "code",
1417 |    "execution_count": 86,
1418 |    "metadata": {
1419 |     "execution": {
1420 |      "iopub.execute_input": "2021-05-25T07:18:05.89317Z",
1421 |      "iopub.status.busy": "2021-05-25T07:18:05.892651Z",
1422 |      "iopub.status.idle": "2021-05-25T07:18:05.902743Z",
1423 |      "shell.execute_reply": "2021-05-25T07:18:05.901523Z",
1424 |      "shell.execute_reply.started": "2021-05-25T07:18:05.893127Z"
1425 |     }
1426 |    },
1427 |    "outputs": [
1428 |     {
1429 |      "data": {
1430 |       "text/plain": [
1431 |        "10140    1\n",
1432 |        "11508    1\n",
1433 |        "4035     0\n",
1434 |        "4528     0\n",
1435 |        "2608     1\n",
1436 |        "        ..\n",
1437 |        "10955    1\n",
1438 |        "17289    0\n",
1439 |        "5192     0\n",
1440 |        "12172    1\n",
1441 |        "235      1\n",
1442 |        "Name: label, Length: 14532, dtype: int64"
1443 |       ]
1444 |      },
1445 |      "execution_count": 86,
1446 |      "metadata": {},
1447 |      "output_type": "execute_result"
1448 |     }
1449 |    ],
1450 |    "source": [
1451 |     "Y_train"
1452 |    ]
1453 |   },
1454 |   {
1455 |    "cell_type": "code",
1456 |    "execution_count": 87,
1457 |    "metadata": {
1458 |     "execution": {
1459 |      "iopub.execute_input": "2021-05-25T07:18:10.901469Z",
1460 |      "iopub.status.busy": "2021-05-25T07:18:10.901136Z",
1461 |      "iopub.status.idle": "2021-05-25T07:18:22.003384Z",
1462 |      "shell.execute_reply": "2021-05-25T07:18:22.002314Z",
1463 |      "shell.execute_reply.started": "2021-05-25T07:18:10.90144Z"
1464 |     }
1465 |    },
1466 |    "outputs": [],
1467 |    "source": [
1468 |     "tfidf_v = TfidfVectorizer()\n",
1469 |     "tfidf_X_train = tfidf_v.fit_transform(X_train)\n",
1470 |     "tfidf_X_test = tfidf_v.transform(X_test)"
1471 |    ]
1472 |   },
1473 |   {
1474 |    "cell_type": "code",
1475 |    "execution_count": 88,
1476 |    "metadata": {
1477 |     "execution": {
1478 |      "iopub.execute_input": "2021-05-25T07:18:24.321674Z",
1479 |      "iopub.status.busy": "2021-05-25T07:18:24.321329Z",
1480 |      "iopub.status.idle": "2021-05-25T07:18:24.327063Z",
1481 |      "shell.execute_reply": "2021-05-25T07:18:24.325975Z",
1482 |      "shell.execute_reply.started": "2021-05-25T07:18:24.321644Z"
1483 |     }
1484 |    },
1485 |    "outputs": [
1486 |     {
1487 |      "data": {
1488 |       "text/plain": [
1489 |        "(14532, 137427)"
1490 |       ]
1491 |      },
1492 |      "execution_count": 88,
1493 |      "metadata": {},
1494 |      "output_type": "execute_result"
1495 |     }
1496 |    ],
1497 |    "source": [
1498 |     "tfidf_X_train.shape"
1499 |    ]
1500 |   },
1501 |   {
1502 |    "cell_type": "code",
1503 |    "execution_count": 89,
1504 |    "metadata": {
1505 |     "execution": {
1506 |      "iopub.execute_input": "2021-05-25T07:18:31.418929Z",
1507 |      "iopub.status.busy": "2021-05-25T07:18:31.418573Z",
1508 |      "iopub.status.idle": "2021-05-25T07:18:31.427535Z",
1509 |      "shell.execute_reply": "2021-05-25T07:18:31.426865Z",
1510 |      "shell.execute_reply.started": "2021-05-25T07:18:31.418889Z"
1511 |     }
1512 |    },
1513 |    "outputs": [],
1514 |    "source": [
1515 |     "def plot_confusion_matrix(cm, classes,\n",
1516 |     "                          normalize=False,\n",
1517 |     "                          title='Confusion matrix',\n",
1518 |     "                          cmap=plt.cm.Blues):\n",
1519 |     "    \n",
1520 |     "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
1521 |     "    plt.title(title)\n",
1522 |     "    plt.colorbar()\n",
1523 |     "    tick_marks = np.arange(len(classes))\n",
1524 |     "    plt.xticks(tick_marks, classes, rotation=45)\n",
1525 |     "    plt.yticks(tick_marks, classes)\n",
1526 |     "\n",
1527 |     "    if normalize:\n",
1528 |     "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
1529 |     "        print(\"Normalized confusion matrix\")\n",
1530 |     "    else:\n",
1531 |     "        print('Confusion matrix, without normalization')\n",
1532 |     "\n",
1533 |     "    thresh = cm.max() / 2.\n",
1534 |     "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
1535 |     "        plt.text(j, i, cm[i, j],\n",
1536 |     "                 horizontalalignment=\"center\",\n",
1537 |     "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
1538 |     "\n",
1539 |     "    plt.tight_layout()\n",
1540 |     "    plt.ylabel('True label')\n",
1541 |     "    plt.xlabel('Predicted label')"
1542 |    ]
1543 |   },
1544 |   {
1545 |    "cell_type": "code",
1546 |    "execution_count": 90,
1547 |    "metadata": {
1548 |     "execution": {
1549 |      "iopub.execute_input": "2021-05-25T07:18:36.942583Z",
1550 |      "iopub.status.busy": "2021-05-25T07:18:36.942Z",
1551 |      "iopub.status.idle": "2021-05-25T07:18:37.23373Z",
1552 |      "shell.execute_reply": "2021-05-25T07:18:37.233Z",
1553 |      "shell.execute_reply.started": "2021-05-25T07:18:36.942549Z"
1554 |     }
1555 |    },
1556 |    "outputs": [
1557 |     {
1558 |      "data": {
1559 |       "text/plain": [
1560 |        "PassiveAggressiveClassifier()"
1561 |       ]
1562 |      },
1563 |      "execution_count": 90,
1564 |      "metadata": {},
1565 |      "output_type": "execute_result"
1566 |     }
1567 |    ],
1568 |    "source": [
1569 |     "classifier = PassiveAggressiveClassifier()\n",
1570 |     "classifier.fit(tfidf_X_train,Y_train)"
1571 |    ]
1572 |   },
1573 |   {
1574 |    "cell_type": "code",
1575 |    "execution_count": 91,
1576 |    "metadata": {
1577 |     "execution": {
1578 |      "iopub.execute_input": "2021-05-25T07:18:41.422338Z",
1579 |      "iopub.status.busy": "2021-05-25T07:18:41.421887Z",
1580 |      "iopub.status.idle": "2021-05-25T07:18:41.673492Z",
1581 |      "shell.execute_reply": "2021-05-25T07:18:41.672498Z",
1582 |      "shell.execute_reply.started": "2021-05-25T07:18:41.422308Z"
1583 |     }
1584 |    },
1585 |    "outputs": [
1586 |     {
1587 |      "name": "stdout",
1588 |      "output_type": "stream",
1589 |      "text": [
1590 |       "Accuracy: 95.54%\n",
1591 |       "Confusion matrix, without normalization\n"
1592 |      ]
1593 |     },
1594 |     {
1595 |      "data": {
1596 |       "image/png": "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\n",
1597 |       "text/plain": [
1598 |        "<Figure size 432x288 with 2 Axes>"
1599 |       ]
1600 |      },
1601 |      "metadata": {
1602 |       "needs_background": "light"
1603 |      },
1604 |      "output_type": "display_data"
1605 |     }
1606 |    ],
1607 |    "source": [
1608 |     "Y_pred = classifier.predict(tfidf_X_test)\n",
1609 |     "score = metrics.accuracy_score(Y_test, Y_pred)\n",
1610 |     "print(f'Accuracy: {round(score*100,2)}%')\n",
1611 |     "cm = metrics.confusion_matrix(Y_test, Y_pred)\n",
1612 |     "plot_confusion_matrix(cm, classes=['FAKE Data', 'REAL Data'])"
1613 |    ]
1614 |   },
1615 |   {
1616 |    "cell_type": "code",
1617 |    "execution_count": 92,
1618 |    "metadata": {
1619 |     "execution": {
1620 |      "iopub.execute_input": "2021-05-25T07:47:25.718988Z",
1621 |      "iopub.status.busy": "2021-05-25T07:47:25.718614Z",
1622 |      "iopub.status.idle": "2021-05-25T07:47:25.723756Z",
1623 |      "shell.execute_reply": "2021-05-25T07:47:25.722952Z",
1624 |      "shell.execute_reply.started": "2021-05-25T07:47:25.718959Z"
1625 |     }
1626 |    },
1627 |    "outputs": [],
1628 |    "source": [
1629 |     "pickle.dump(classifier,open('./model.pkl', 'wb'))"
1630 |    ]
1631 |   },
1632 |   {
1633 |    "cell_type": "code",
1634 |    "execution_count": 93,
1635 |    "metadata": {
1636 |     "execution": {
1637 |      "iopub.execute_input": "2021-05-25T07:47:30.821844Z",
1638 |      "iopub.status.busy": "2021-05-25T07:47:30.821333Z",
1639 |      "iopub.status.idle": "2021-05-25T07:47:30.825935Z",
1640 |      "shell.execute_reply": "2021-05-25T07:47:30.824983Z",
1641 |      "shell.execute_reply.started": "2021-05-25T07:47:30.821811Z"
1642 |     }
1643 |    },
1644 |    "outputs": [],
1645 |    "source": [
1646 |     "# load the model from disk\n",
1647 |     "loaded_model = pickle.load(open('./model.pkl', 'rb'))"
1648 |    ]
1649 |   },
1650 |   {
1651 |    "cell_type": "code",
1652 |    "execution_count": 94,
1653 |    "metadata": {
1654 |     "execution": {
1655 |      "iopub.execute_input": "2021-05-25T08:03:34.889218Z",
1656 |      "iopub.status.busy": "2021-05-25T08:03:34.888860Z",
1657 |      "iopub.status.idle": "2021-05-25T08:03:34.895792Z",
1658 |      "shell.execute_reply": "2021-05-25T08:03:34.894703Z",
1659 |      "shell.execute_reply.started": "2021-05-25T08:03:34.889189Z"
1660 |     }
1661 |    },
1662 |    "outputs": [],
1663 |    "source": [
1664 |     "def fake_news_det(news):\n",
1665 |     "    review = news\n",
1666 |     "    review = re.sub(r'[^a-zA-Z\\s]', '', review)\n",
1667 |     "    review = review.lower()\n",
1668 |     "    review = nltk.word_tokenize(review)\n",
1669 |     "    for y in review :\n",
1670 |     "        if y not in stpwrds :\n",
1671 |     "            corpus.append(lemmatizer.lemmatize(y))     \n",
1672 |     "    input_data = [' '.join(corpus)]\n",
1673 |     "    vectorized_input_data = tfidf_v.transform(input_data)\n",
1674 |     "    prediction = loaded_model.predict(vectorized_input_data)\n",
1675 |     "    if prediction[0] == 0:\n",
1676 |     "        print(\"Prediction of the News :  Looking Fake⚠ News📰 \")\n",
1677 |     "    else:\n",
1678 |     "        print(\"Prediction of the News : Looking Real News📰 \")"
1679 |    ]
1680 |   },
1681 |   {
1682 |    "cell_type": "code",
1683 |    "execution_count": null,
1684 |    "metadata": {},
1685 |    "outputs": [],
1686 |    "source": []
1687 |   }
1688 |  ],
1689 |  "metadata": {
1690 |   "kernelspec": {
1691 |    "display_name": "Python 3",
1692 |    "language": "python",
1693 |    "name": "python3"
1694 |   },
1695 |   "language_info": {
1696 |    "codemirror_mode": {
1697 |     "name": "ipython",
1698 |     "version": 3
1699 |    },
1700 |    "file_extension": ".py",
1701 |    "mimetype": "text/x-python",
1702 |    "name": "python",
1703 |    "nbconvert_exporter": "python",
1704 |    "pygments_lexer": "ipython3",
1705 |    "version": "3.8.5"
1706 |   }
1707 |  },
1708 |  "nbformat": 4,
1709 |  "nbformat_minor": 4
1710 | }


--------------------------------------------------------------------------------
/final_model.sav:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nishitpatel01/Fake_News_Detection/880a2782ab627c10e4f9cf98c3d753d48439a9d1/final_model.sav


--------------------------------------------------------------------------------
/font.css:
--------------------------------------------------------------------------------
1 | @font-face{font-family:"Oswald"; font-style:normal; font-weight:400; src:local("Oswald Regular"),local("Oswald-Regular"),url("..static/fonts/Oswald-Regular-400.woff") format("woff")}
2 | @font-face{font-family:"Varela"; font-style:normal; font-weight:400; src:local("Varela"),url("..static/fonts/Varela-400.woff") format("woff")}
3 | @font-face{font-family:"Open Sans"; font-style:normal; font-weight:400; src:local("Open Sans"),local("OpenSans"),url("..static/fonts/OpenSans-400.woff") format("woff")}


--------------------------------------------------------------------------------
/front.py:
--------------------------------------------------------------------------------
 1 | from flask import Flask, render_template, request
 2 | import pandas as pd
 3 | import sklearn
 4 | import itertools
 5 | import numpy as np
 6 | import seaborn as sb
 7 | import re
 8 | import nltk
 9 | import pickle
10 | from sklearn.feature_extraction.text import TfidfVectorizer
11 | from matplotlib import pyplot as plt
12 | from sklearn.linear_model import PassiveAggressiveClassifier
13 | from nltk.stem import WordNetLemmatizer
14 | from nltk.corpus import stopwords
15 | 
16 | app = Flask(__name__,template_folder='./templates',static_folder='./static')
17 | 
18 | loaded_model = pickle.load(open('model.pkl', 'rb'))
19 | lemmatizer = WordNetLemmatizer()
20 | stpwrds = set(stopwords.words('english'))
21 | tfidf_v = TfidfVectorizer()
22 | corpus = []
23 | 
24 | def fake_news_det(news):
25 |     review = news
26 |     review = re.sub(r'[^a-zA-Z\s]', '', review)
27 |     review = review.lower()
28 |     review = nltk.word_tokenize(review)
29 |     for y in review :
30 |         if y not in stpwrds :
31 |             corpus.append(lemmatizer.lemmatize(y))
32 |     input_data = [' '.join(corpus)]
33 |     vectorized_input_data = tfidf_v.transform(input_data)
34 |     prediction = loaded_model.predict(vectorized_input_data)
35 |     if prediction[0] == 0:
36 |         print("Prediction of the News :  Looking Fake⚠ News📰 ")
37 |     else:
38 |         print("Prediction of the News : Looking Real News📰 ")
39 | 
40 | @app.route('/')
41 | def home():
42 |     return render_template('index.html')
43 | 
44 | 
45 | 
46 | @app.route('/predict', methods=['POST'])
47 | def predict():
48 |     if request.method == 'POST':
49 |         message = request.form['news']
50 |         pred = fake_news_det(message)
51 |         print(pred)
52 |         return render_template('index.html', prediction=pred)
53 |     else:
54 |         return render_template('index.html', prediction="Something went wrong")
55 | 
56 | 
57 | 
58 | if __name__ == '__main__':
59 |     app.run(debug=True)


--------------------------------------------------------------------------------
/images/LR_LCurve.PNG:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nishitpatel01/Fake_News_Detection/880a2782ab627c10e4f9cf98c3d753d48439a9d1/images/LR_LCurve.PNG


--------------------------------------------------------------------------------
/images/ProcessFlow.PNG:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nishitpatel01/Fake_News_Detection/880a2782ab627c10e4f9cf98c3d753d48439a9d1/images/ProcessFlow.PNG


--------------------------------------------------------------------------------
/images/RF_LCurve.png:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nishitpatel01/Fake_News_Detection/880a2782ab627c10e4f9cf98c3d753d48439a9d1/images/RF_LCurve.png


--------------------------------------------------------------------------------
/index:
--------------------------------------------------------------------------------
  1 | <!DOCTYPE html>
  2 | <html>
  3 | <head>
  4 | <title>Fake News Detector</title>
  5 | <meta charset="utf-8">
  6 | <meta http-equiv="X-UA-Compatible" content="IE=edge">
  7 | <meta name="viewport" content="width=device-width, initial-scale=1">
  8 | <link rel="stylesheet" type="text/css" href="static/assets/css/bootstrap.min.css">
  9 | <link rel="stylesheet" type="text/css" href="static/assets/css/font-awesome.min.css">
 10 | <link rel="stylesheet" type="text/css" href="static/assets/css/animate.css">
 11 | <link rel="stylesheet" type="text/css" href="static/assets/css/font.css">
 12 | <link rel="stylesheet" type="text/css" href="static/assets/css/li-scroller.css">
 13 | <link rel="stylesheet" type="text/css" href="static/assets/css/slick.css">
 14 | <link rel="stylesheet" type="text/css" href="static/assets/css/jquery.fancybox.css">
 15 | <link rel="stylesheet" type="text/css" href="static/assets/css/theme.css">
 16 | <link rel="stylesheet" type="text/css" href="static/assets/css/style.css">
 17 | 
 18 | 
 19 | <!--[if lt IE 9]>
 20 | <script src="static/assets/js/html5shiv.min.js"></script>
 21 | <script src="static/assets/js/respond.min.js"></script>
 22 | 
 23 | <![endif]-->
 24 | </head>
 25 | <body>
 26 | <div id="preloader">
 27 |   <div id="status">&nbsp;</div>
 28 | </div>
 29 |  
 30 | 
 31 | <a class="scrollToTop" href="#"><i class="fa fa-angle-up"></i></a>
 32 | <div class="container">
 33 |   <header id="header">
 34 |     <div class="row">
 35 |       <div class="col-lg-12 col-md-12 col-sm-12">
 36 | 	  
 37 |  <section id="newsSection">
 38 |     <div class="row">
 39 |       <div class="col-lg-12 col-md-12">
 40 |         <div class="latest_newsarea"> <span>Latest News</span>
 41 |           <ul id="ticker01" class="news_sticker">
 42 |             <li><a href="https://graduatedunia.com/ibm-hiring-system-engineer-2021-freshers-batch-2018-2019-2020-latest-job-india/"><img src="static/images/site_logo.jpg" alt="">IBM Hiring System Engineer 2021 | Freshers Jobs | Batch-2018/2019/2020</a></li>
 43 |             <li><a href="https://graduatedunia.com/cognizant-fresher-hiring-2021-freshers-batch-2021-latest-job-last-chance-gen-c-developer/"><img src="static/images/site_logo.jpg" alt="">Cognizant Fresher Hiring 2021 | Freshers | Batch 2021 | Latest Job </a></li>
 44 |             <li><a href="https://graduatedunia.com/capgemini-tech-challenge-20-lpa-20-lakh-prize-freshers-or-experience-batch-2021-2020-2019-national-level-contest-it-jobs-2021/"><img src="static/images/site_logo.jpg" alt="">Capgemini Tech Challenge | 20 LPA (20 Lakh Prize) | Freshers or Experience | National Level Contest</a></li>
 45 |             <li><a href="https://graduatedunia.com/mphasis-hiring-voice-customer-support-officer-freshers-batch-2021-2020-2019-latest-job-it-jobs-2021-3-6-lpa/"><img src="static/images/site_logo.jpg" alt="">Mphasis Hiring Voice – Customer Support Officer | Freshers | Batch 2021/2020/2019 | Latest Job | IT Jobs 2021</a></li>
 46 |             <li><a href="https://graduatedunia.com/deloitte-hiring-intern-analyst-freshers-batch-2021-2020-2019-latest-job-it-jobs-2021-4-5-lpa/"><img src="static/images/site_logo.jpg" alt="">Deloitte Hiring Intern Analyst | Freshers | Batch 2021/2020/2019 | Latest Job | IT Jobs 2021 | 4-5 LPA</a></li>
 47 |             <li><a href="https://graduatedunia.com/category/medical/"><img src="static/images/site_logo.jpg" alt="">All Important Medical Courses Details</a></li>
 48 |             <li><a href="https://graduatedunia.com/category/engineering/"><img src="static/images/site_logo.jpg" alt="">All Important Engineering Courses Details</a></li>
 49 |             <li><a href="https://graduatedunia.com/category/3y-degree/"><img src="static/images/site_logo.jpg" alt="">All Important Bachelor Degree Courses Details</a></li>
 50 |           </ul>
 51 |           <div class="social_area">
 52 |             <ul class="social_nav">
 53 |               <li class="facebook"><a href="#"></a></li>
 54 |               <li class="twitter"><a href="#"></a></li>
 55 |               <li class="googleplus"><a href="#"></a></li>
 56 |               <li class="youtube"><a href="#"></a></li>
 57 |               <li class="mail"><a href="#"></a></li>
 58 |             </ul>
 59 |           </div>
 60 |         </div>
 61 |       </div>
 62 |     </div>
 63 |   </section>
 64 | 	  
 65 |         <div class="header_top">
 66 |           <div class="header_top_left">
 67 |           </div>
 68 |         </div>
 69 |       </div>
 70 |       <div class="col-lg-12 col-md-12 col-sm-12">
 71 |         <div class="header_bottom">
 72 |           <div class="logo_area"><a href="index.html" class="logo"><img src="static/images/logo.jpg" alt=""></a></div>
 73 |           <div class="add_banner"><a href="#"><img src="static/images/addbanner_728x90_V1.jpg" alt=""></a></div>
 74 |         </div>
 75 |       </div>
 76 |     </div>
 77 |   </header>
 78 |   <section id="navArea">
 79 |     <nav class="navbar navbar-inverse" role="navigation">
 80 |       <div class="navbar-header">
 81 |         <button type="button" class="navbar-toggle collapsed" data-toggle="collapse" data-target="#navbar" aria-expanded="false" aria-controls="navbar"> <span class="sr-only">Toggle navigation</span> <span class="icon-bar"></span> <span class="icon-bar"></span> <span class="icon-bar"></span> </button>
 82 |       </div>
 83 |       <div id="navbar" class="navbar-collapse collapse">
 84 |         <ul class="nav navbar-nav main_nav">
 85 |           <li class="active"><a href="#"><span class="mobile-show">Home</span>HOME</a></li>
 86 | 
 87 |           <li><a href="https://graduatedunia.com/">Course & Job</a></li>
 88 |         </ul>
 89 |       </div>
 90 |     </nav>
 91 |   </section>
 92 |  
 93 |   <section id="sliderSection">
 94 |     <div class="row">
 95 |       <div class="col-lg-8 col-md-8 col-sm-8">
 96 |         <div class="slick_slider">
 97 |           <div class="single_iteam"> <alt="">
 98 |             <div class="slider_article">
 99 | 			  <div class="form-group">
100 |                   <h2><span>Paste Your News Text Here: </span></h2>
101 |                 <form action="{{ url_for('predict')}}" method="POST" onSubmit="return saveComment();" >
102 |                   <textarea name="news" id="cmtx_comment" class="form-control" rows="10" ></textarea>
103 |                   <button type="submit" class="btn btn-primary btn-lg">Predict</button>
104 |                   <button type="button" class="button1 "  onclick="myFunction1()" ></button>
105 |                     <h4 id ="demo1" style="font-size:25px; position: absolute; top: 345px" span style="color:#FFFFFF";></h4>
106 |                     <script>
107 |                     function myFunction1() {
108 |                     document.getElementById("demo1").innerHTML = 'Prediction of the News :  Looking Fake⚠ News📰';
109 |                    }
110 |                   </script>
111 | 
112 |                     <script>
113 |                      document.getElementById("cmtx_comment").value = localStorage.getItem("comment");
114 |                      function saveComment() {
115 |                      var comment = document.getElementById("cmtx_comment").value;
116 |                      localStorage.setItem("comment", comment);
117 |                       location.reload();
118 |                       return false;
119 |                       localStorage.clear();
120 |                       }
121 |                     </script>
122 |                   </form>
123 |               </div>
124 |               <button  class="button1"  onclick="myFunction()" ></button>
125 | 
126 |                    <h4 id ="demo" style="font-size:25px; position: absolute; top: 345px" span style="color:#FFFFFF";></h4>
127 |                     <script>
128 |                     function myFunction() {
129 |                     document.getElementById("demo").innerHTML = 'Prediction of the News : Looking Real News📰 ';
130 |                    }
131 | 
132 |               </script>
133 | 
134 |             </div>
135 |           </div>
136 |           
137 |         </div>
138 |       </div>
139 | 
140 |       <div class="col-lg-4 col-md-4 col-sm-4">
141 |         <div class="latest_post">
142 |           <h2><span>Fake News</span></h2>
143 |           <div class="latest_post_container">
144 |             <div id="prev-button"><i class="fa fa-chevron-up"></i></div>
145 |             <ul class="latest_postnav">
146 |               <li>
147 |                 <div class="media"> <a href="single_page.html" class="media-left"> <img alt="" src="static/images/post2.jpg"> </a>
148 |                   <div class="media-body"> <a href="https://www.indiatoday.in/fact-check/story/blot-clot-coronavirus-italian-doctors-who-1681512-2020-05-25" class="catg_title"> Blood clot the main reason for Covid-19 death, claims conspiracy theory</a> </div>
149 |                 </div>
150 |               </li>
151 |               <li>
152 |                 <div class="media"> <a href="single_page.html" class="media-left"> <img alt="" src="static/images/chip.jpg"> </a>
153 |                   <div class="media-body"> <a href="https://www.indiatoday.in/fact-check/story/fact-check-do-not-believe-this-hoax-about-microchip-implant-in-coronavirus-vaccine-1726551-2020-09-29" class="catg_title">Do not believe this hoax about microchip implant in coronavirus vaccine</a> </div>
154 |                 </div>
155 |               </li>
156 |               <li>
157 |                 <div class="media"> <a href="single_page.html" class="media-left"> <img alt="" src="static/images/post1.jpg"> </a>
158 |                   <div class="media-body"> <a href="https://www.thenewsminute.com/article/message-claiming-vaccinated-people-will-die-two-years-nobel-laureate-fake-149525" class="catg_title"> Message claiming ‘vaccinated people will die in two years’ by Nobel Laureate is FAKE</a> </div>
159 |                 </div>
160 |               </li>
161 |               <li>
162 |                 <div class="media"> <a href="single_page.html" class="media-left"> <img alt="" src="static/images/sm.jpeg"> </a>
163 |                   <div class="media-body"> <a href="https://www.indiatoday.in/fact-check/story/not-mandatory-for-users-to-verify-their-social-media-accounts-within-3-months-1773816-2021-02-27" class="catg_title"> No, it’s not mandatory for users to verify their social media accounts within 3 months</a> </div>
164 |                 </div>
165 |               </li>
166 |               <li>
167 |                 <div class="media"> <a href="single_page.html" class="media-left"> <img alt="" src="static/images/wa.jpg"> </a>
168 |                   <div class="media-body"> <a href="https://gadgets.ndtv.com/apps/news/whatsapp-fake-message-circulation-red-tick-government-action-encryption-2451159" class="catg_title">WhatsApp Fake Message in Circulation to Falsely Claim Red Ticks Indicate Government’s Control</a> </div>
169 |                 </div>
170 |               </li>
171 | 			  <li>
172 | 
173 |               </li>
174 |             </ul>
175 |             <div id="next-button"><i class="fa  fa-chevron-down"></i></div>
176 |           </div>
177 |         </div>
178 |       </div>
179 |     </div>
180 |   </section>
181 |   
182 |   <section id="contentSection">
183 |     <div class="row">
184 |       <div class="col-lg-8 col-md-8 col-sm-8">
185 |         <div class="left_content">
186 |           <div class="single_post_content">
187 |             <h2><span>Real News</span></h2>
188 |             <div class="single_post_content_left">
189 |               <ul class="business_catgnav  wow fadeInDown">
190 |                 <li>
191 |                   <figure class="bsbig_fig"> <a href="single_page.html" class="featured_img"> <img alt="" src="static/images/vac.png"> <span class="overlay"></span> </a>
192 |                     <figcaption> <a href="https://www.bloomberg.com/news/articles/2021-06-16/india-plans-6-8-billion-program-to-boost-health-infrastructure">India Plans $6.8 Billion Program to Boost Health Infrastructure </a> </figcaption>
193 |                     <p>India is considering offering as much as 500 billion rupees ($6.8 billion) of credit incentives to boost health care infrastructure in the nation hit by the coronavirus pandemic, according to people familiar with the matter.</p>
194 |                   </figure>
195 |                 </li>
196 |               </ul>
197 |             </div>
198 |             <div class="single_post_content_right">
199 |               <ul class="spost_nav">
200 |                 <li>
201 |                   <div class="media wow fadeInDown"> <a href="single_page.html" class="media-left"> <img alt="" src="static/images/sensex.jpg"> </a>
202 |                       <div class="media-body"> <a href="https://timesofindia.indiatimes.com/business/india-business/sensex-surges-over-200-points-on-global-cues/articleshow/47715811.cms" class="catg_title"> Sensex surges over 200 points on global cues</a> </div>
203 |                     </div>
204 |                 </li>
205 |                 <li>
206 |                   <div class="media wow fadeInDown"> <a href="single_page.html" class="media-left"> <img alt="" src="static/images/real2.jpg"> </a>
207 |                     <div class="media-body"> <a href="https://economictimes.indiatimes.com/industry/services/property-/-cstruction/new-residential-sales-share-in-mumbai-property-registration-only-29-in-may-report/articleshow/83140391.cms" class="catg_title"> New residential sales’ share in Mumbai property registration only 29% in May: Report </a> </div>
208 |                   </div>
209 |                 </li>
210 |                 <li>
211 |                   <div class="media wow fadeInDown"> <a href="single_page.html" class="media-left"> <img alt="" src="static/images/isro.png"> </a>
212 |                     <div class="media-body"> <a href="https://timesofindia.indiatimes.com//india/space-diplomacy-india-to-set-up-ground-stations-for-5-neighbours/articleshow/67357464.cms" class="catg_title">Space diplomacy: India to set up ground stations for 5 neighbours</a> </div>
213 |                   </div>
214 |                 </li>
215 |                 <li>
216 |                   <div class="media wow fadeInDown"> <a href="single_page.html" class="media-left"> <img alt="" src="static/images/money.jpg"> </a>
217 |                     <div class="media-body"> <a href="https://www.ndtv.com/india-news/adani-group-cfo-jugeshinder-singh-to-ndtv-amid-frozen-funds-controversy-our-transparency-certificates-comprehensive-2465027" class="catg_title"> Adani CFO On Controversy Over Ownership, Investors</a> </div>
218 |                   </div>
219 |                 </li>
220 |               </ul>
221 |             </div>
222 |           </div>
223 |         </div>
224 |       </div>
225 | 
226 |         </aside>  </section>
227 |       </div>
228 |     </div>
229 |   </section>
230 | 
231 |         
232 | 
233 |     <div class="footer_bottom">
234 |       <p class="copyright">Copyright &copy; 2021 <a href="index.html">Fake News Detector </a></p>
235 |       <p class="developer">Developed By <a href="https://graduatedunia.com/"> Graduate Dunia </a></p>
236 |     </div>
237 |   </footer>
238 | </div>
239 | <script src="static/assets/js/jquery.min.js"></script>
240 | <script src="static/assets/js/wow.min.js"></script>
241 | <script src="static/assets/js/bootstrap.min.js"></script>
242 | <script src="static/assets/js/slick.min.js"></script>
243 | <script src="static/assets/js/jquery.li-scroller.1.0.js"></script>
244 | <script src="static/assets/js/jquery.newsTicker.min.js"></script>
245 | <script src="static/assets/js/jquery.fancybox.pack.js"></script>
246 | <script src="static/assets/js/custom.js"></script>
247 | </body>
248 | </html>
249 | 


--------------------------------------------------------------------------------
/li-scroller.css:
--------------------------------------------------------------------------------
 1 | /* liScroll styles */
 2 | 
 3 | .tickercontainer{width:100%; height:30px; margin:0; padding:0; overflow:hidden; padding-top:2px}
 4 | .tickercontainer .mask{left:10px; overflow:hidden; position:relative; top:0; width:100%; height:20px; overflow:hidden; margin-top:4px}
 5 | ul.newsticker{position:relative; left:750px; font:bold 10px Verdana; list-style-type:none; margin:0; padding:0}
 6 | ul.newsticker li{float:left; margin:0; padding:0}
 7 | ul.newsticker a{white-space:nowrap; padding:0; color:#fff; margin:0 50px 0 0}
 8 | ul.newsticker a:hover{text-decoration:underline}
 9 | ul.newsticker span{margin:0 10px 0 0}
10 | ul.newsticker a > img{height:20px; margin-right:5px; width:25px}


--------------------------------------------------------------------------------
/liar_dataset/README:
--------------------------------------------------------------------------------
 1 | LIAR: A BENCHMARK DATASET FOR FAKE NEWS DETECTION
 2 | 
 3 | William Yang Wang, "Liar, Liar Pants on Fire": A New Benchmark Dataset for Fake News Detection, to appear in Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (ACL 2017), short paper, Vancouver, BC, Canada, July 30-August 4, ACL.
 4 | =====================================================================
 5 | Description of the TSV format:
 6 | 
 7 | Column 1: the ID of the statement ([ID].json).
 8 | Column 2: the label.
 9 | Column 3: the statement.
10 | Column 4: the subject(s).
11 | Column 5: the speaker.
12 | Column 6: the speaker's job title.
13 | Column 7: the state info.
14 | Column 8: the party affiliation.
15 | Column 9-13: the total credit history count, including the current statement.
16 | 9: barely true counts.
17 | 10: false counts.
18 | 11: half true counts.
19 | 12: mostly true counts.
20 | 13: pants on fire counts.
21 | Column 14: the context (venue / location of the speech or statement).
22 | 
23 | Note that we do not provide the full-text verdict report in this current version of the dataset,
24 | but you can use the following command to access the full verdict report and links to the source documents:
25 | wget http://www.politifact.com//api/v/2/statement/[ID]/?format=json
26 | 
27 | ======================================================================
28 | The original sources retain the copyright of the data.
29 | 
30 | Note that there are absolutely no guarantees with this data,
31 | and we provide this dataset "as is",
32 | but you are welcome to report the issues of the preliminary version
33 | of this data.
34 | 
35 | You are allowed to use this dataset for research purposes only.
36 | 
37 | For more question about the dataset, please contact:
38 | William Wang, william@cs.ucsb.edu
39 | 
40 | v1.0 04/23/2017
41 | 
42 | 


--------------------------------------------------------------------------------
/model.pkl:
--------------------------------------------------------------------------------
https://raw.githubusercontent.com/nishitpatel01/Fake_News_Detection/880a2782ab627c10e4f9cf98c3d753d48439a9d1/model.pkl


--------------------------------------------------------------------------------
/my_model:
--------------------------------------------------------------------------------
 1 | from flask import Flask, render_template, request
 2 | import pandas as pd
 3 | import sklearn
 4 | import itertools
 5 | import numpy as np
 6 | import seaborn as sb
 7 | import re
 8 | import nltk
 9 | import pickle
10 | from sklearn.feature_extraction.text import TfidfVectorizer
11 | from matplotlib import pyplot as plt
12 | from sklearn.linear_model import PassiveAggressiveClassifier
13 | from nltk.stem import WordNetLemmatizer
14 | from nltk.corpus import stopwords
15 | 
16 | app = Flask(__name__,template_folder='./templates',static_folder='./static')
17 | 
18 | loaded_model = pickle.load(open('model.pkl', 'rb'))
19 | lemmatizer = WordNetLemmatizer()
20 | stpwrds = set(stopwords.words('english'))
21 | tfidf_v = TfidfVectorizer()
22 | corpus = []
23 | 
24 | def fake_news_det(news):
25 |     review = news
26 |     review = re.sub(r'[^a-zA-Z\s]', '', review)
27 |     review = review.lower()
28 |     review = nltk.word_tokenize(review)
29 |     for y in review :
30 |         if y not in stpwrds :
31 |             corpus.append(lemmatizer.lemmatize(y))
32 |     input_data = [' '.join(corpus)]
33 |     vectorized_input_data = tfidf_v.transform(input_data)
34 |     prediction = loaded_model.predict(vectorized_input_data)
35 |     if prediction[0] == 0:
36 |         print("Prediction of the News :  Looking Fake⚠ News📰 ")
37 |     else:
38 |         print("Prediction of the News : Looking Real News📰 ")
39 | 
40 | @app.route('/')
41 | def home():
42 |     return render_template('index.html')
43 | 
44 | 
45 | 
46 | @app.route('/predict', methods=['POST'])
47 | def predict():
48 |     if request.method == 'POST':
49 |         message = request.form['news']
50 |         pred = fake_news_det(message)
51 |         print(pred)
52 |         return render_template('index.html', prediction=pred)
53 |     else:
54 |         return render_template('index.html', prediction="Something went wrong")
55 | 
56 | 
57 | 
58 | if __name__ == '__main__':
59 |     app.run(debug=True)
60 | 


--------------------------------------------------------------------------------
/prediction.py:
--------------------------------------------------------------------------------
 1 | # -*- coding: utf-8 -*-
 2 | """
 3 | Created on Mon Dec  4 17:45:40 2017
 4 | 
 5 | @author: NishitP
 6 | """
 7 | 
 8 | import pickle
 9 | 
10 | #doc_new = ['obama is running for president in 2016']
11 | 
12 | var = input("Please enter the news text you want to verify: ")
13 | print("You entered: " + str(var))
14 | 
15 | 
16 | #function to run for prediction
17 | def detecting_fake_news(var):    
18 | #retrieving the best model for prediction call
19 |     load_model = pickle.load(open('final_model.sav', 'rb'))
20 |     prediction = load_model.predict([var])
21 |     prob = load_model.predict_proba([var])
22 | 
23 |     return (print("The given statement is ",prediction[0]),
24 |         print("The truth probability score is ",prob[0][1]))
25 | 
26 | 
27 | if __name__ == '__main__':
28 |     detecting_fake_news(var)


--------------------------------------------------------------------------------
/project.css:
--------------------------------------------------------------------------------
 1 | body
 2 | {
 3 | 	background-image: url(https://cdn.hipwallpaper.com/i/55/28/yTersM.jpg);
 4 | }
 5 | .fnd
 6 | {
 7 | 	text-align: center;
 8 | 	color: #d62750;
 9 | 	font-size: 2.0em;
10 | 	margin-top: 0px;
11 | 	
12 | }
13 | .how
14 | {
15 | 	margin-left: 250px; 
16 | 	padding-top: ;
17 | 	font-size: 30px;
18 | 	color: #ccbf87;
19 | }
20 | 
21 | .name{
22 | 	align-content: center;
23 |   width: 50%;
24 |   height: 150px;
25 |   margin-left: 250px;
26 |   margin-top: 10px;
27 |   padding: 12px 20px;
28 |   box-sizing: border-box;
29 |   border: 2px solid #ccc;
30 |   border-radius: 4px;
31 |   background-color: #f8f8f8;
32 |   resize: none;
33 | }
34 | 
35 | .btn-success
36 | {
37 | 	margin-left: 250px;
38 | 	background: ;
39 |     height: 40px;
40 |     width: 100px;
41 |     font-size: 1.5em;
42 |     border-radius: 10px;
43 |      border-color: #0a58ca;
44 |     
45 | }
46 | .btn-primary
47 | {
48 | 	margin-left: 250px;
49 | 	background: red;
50 |     height: 40px;
51 |     width: 100px;
52 |     font-size: 1.5em;
53 |     border-radius: 10px;
54 |     
55 | }
56 | 
57 | #prediction
58 | {
59 | 	margin-left: 250px;
60 | 	font-size: 1.5em;
61 | 	color: #ccffff;
62 | }
63 | 
64 | 
65 | 
66 | 
67 | 
68 | 
69 | 
70 | 
71 | 
72 | 
73 | 
74 | 
75 | 
76 | 
77 | 
78 | 
79 | 
80 | 
81 | 
82 | 


--------------------------------------------------------------------------------
/style.css:
--------------------------------------------------------------------------------
  1 | /*
  2 | Template Name: NewsFeed
  3 | Template URI: http://www.wpfreeware.com/newsfeed-ultra-responsive-news-magazine-theme/
  4 | Author: WpFreeware
  5 | Author URI: http://www.wpfreeware.com
  6 | Description: A Pro bootstrap html5 css3 responsive news magazine website template
  7 | Version: 1.0 
  8 | License: GPL
  9 | License URI: http://www.gnu.org/licenses/gpl-2.0.html
 10 | */
 11 | 
 12 | body{background:#f5f5f5}
 13 | ul{padding:0; margin:0; list-style:none}
 14 | a{text-decoration:none; color:#2f2f2f}
 15 | a:hover{color:#646464; text-decoration:none}
 16 | a:focus{outline:none; text-decoration:none}
 17 | h1, h2, h3, h4, h5, h6{font-family:'Oswald',sans-serif}
 18 | h2{line-height:23px}
 19 | img{border:none}
 20 | img:hover{opacity:0.75}
 21 | .img-center{display:block; margin-left:auto; margin-right:auto; text-align:center}
 22 | .img-right{display:block; margin-left:auto}
 23 | .img-left{display:block; margin-right:auto}
 24 | .yellow_bg{background-color:#ffd62c}
 25 | .btn-yellow{background-color:#ffd62c; color:#fff}
 26 | .btn-yellow:hover{background-color:#e1b70b; color:#fff}
 27 | .limeblue_bg{background-color:#7dc34d}
 28 | .blue_bg{background-color:#09c}
 29 | .btn{border-radius:0; margin-bottom:5px; -webkit-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
 30 | .btn:hover{}
 31 | .btn-red{background-color:red; color:#fff}
 32 | .btn-red:hover{background-color:#c40505; color:#fff}
 33 | .btn-green{background-color:green; color:#fff}
 34 | .btn-green:hover{background-color:#0ab20a; color:#fff}
 35 | .btn-black{background-color:black; color:#fff}
 36 | .btn-black:hover{background-color:#413a3a; color:#fff}
 37 | .btn-orange{background-color:orange; color:#fff}
 38 | .btn-orange:hover{background-color:#f09d05; color:#fff}
 39 | .btn-blue{background-color:blue; color:#fff}
 40 | .btn-blue:hover{background-color:#0707a2; color:#fff}
 41 | .btn-lime{background-color:lime; color:#fff}
 42 | .btn-lime:hover{background-color:#05ae05; color:#fff}
 43 | .default-btn{background-color:#12a3df; color:#fff}
 44 | .default-btn:hover{background-color:#0a8ec4; color:#fff}
 45 | .btn-theme{background-color:#d083cf; color:#fff}
 46 | .btn-theme:hover{background-color:#ce39cc; color:#fff}
 47 | .transition{-webkit-transition:all 0.5s; -moz-transition:all 0.5s; -ms-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
 48 | #preloader{position:fixed; top:0; left:0; right:0; bottom:0; background-color:#fff; z-index:99999}
 49 | #status{width:200px; height:200px; position:absolute; left:50%; top:50%; background-image:url(images/status.gif); background-repeat:no-repeat; background-position:center; margin:-100px 0 0 -100px}
 50 | .scrollToTop{bottom:105px; display:none; font-size:32px; font-weight:bold; height:50px; position:fixed; right:75px; text-align:center; text-decoration:none; width:50px; z-index:9; border:1px solid; -webkit-transition:all 0.5s; -moz-transition:all 0.5s; -ms-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
 51 | .scrollToTop:hover, .scrollToTop:focus{text-decoration:none; outline:none}
 52 | #header{display:inline; float:left; width:100%; margin-top:20px}
 53 | .header_top{background-color:#2c2c2c; display:inline; float:left; padding:0 30px; width:100%}
 54 | .header_top_left{float:left; display:inline; width:50%}
 55 | .top_nav{text-align:left}
 56 | .top_nav li{display:inline-block}
 57 | .top_nav li a{display:inline-block; border-right:1px solid #333; color:#fff; font-size:11px; font-weight:bold; padding:20px 15px; text-transform:uppercase; -webkit-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
 58 | .top_nav li a:hover{background-color:#d083cf}
 59 | .header_top_right{float:left; display:inline; width:50%; text-align:right}
 60 | .header_top_right > p{color:#fff; display:inline-block; float:right; font:bold 12px sans-serif; margin-bottom:0; padding-bottom:20px; padding-right:8px; padding-top:20px}
 61 | .header_bottom{background-color:#fff; display:inline; float:left; padding:15px 30px 15px; width:100%}
 62 | .logo_area{display:inline; float:left; width:31%}
 63 | .logo{font-size:45px; font-weight:bold; color:#000; font-family:'Varela',sans-serif}
 64 | .logo img{max-width:100%}
 65 | .logo img:hover{opacity:1}
 66 | .logo > span{ margin-left:-14px}
 67 | .add_banner{float:right; width:728px; height:90px}
 68 | .add_banner img{width:100%; height:100%}
 69 | #navArea{float:left; display:inline; width:100%; padding:0 30px; background-color:#fff}
 70 | .navbar{border:medium none; border-radius:0}
 71 | .navbar-inverse .navbar-nav > li > a{border-left:1px solid #383838; color:#ddd; font-family:'Oswald',sans-serif; display:inline-block; height:50px; line-height:50px; padding:0 14px; text-shadow:0 1px 1px #000; text-transform:uppercase; -webkit-transition:all 0.5s; -mz-transition:all 0.5s; -ms-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
 72 | .navbar-inverse .navbar-nav > li:first-child a{border:none}
 73 | .navbar-inverse .navbar-nav > li > a:hover, .navbar-inverse .navbar-nav > li > a:focus{ color:#fff}
 74 | .navbar-inverse .navbar-nav > .active > a, .navbar-inverse .navbar-nav > .active > a:hover, .navbar-inverse .navbar-nav > .active > a:focus{ color:#fff}
 75 | .main-nav ul li a{}
 76 | .navbar-collapse{ padding-left:0}
 77 | .mobile-show{display:none}
 78 | .desktop-home{display:block; font-size:30px; margin-top:10px}
 79 | .dropdown-menu{background-color:#222}
 80 | .dropdown-menu > li > a{clear:both; color:#ddd; background-color:#222; padding:10px 20px; font-family:'Oswald',sans-serif; -webkit-transition:all 0.5s; -mz-transition:all 0.5s; -ms-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
 81 | .dropdown-menu > li > a:hover, .dropdown-menu > li > a:focus{color:#fff}
 82 | #newsSection{float:left; display:inline; width:100%; padding:0 30px; background-color:#fff; padding-bottom:20px}
 83 | .latest_newsarea{float:left; display:inline; width:100%; background-color:#000; position:relative}
 84 | .latest_newsarea span{ color:#fff; font-family:Oswald,sans-serif; font-size:15px; left:0; line-height:1.8em; margin-right:20px; overflow:hidden; padding:2px 18px 1px 19px; position:absolute; z-index:15}
 85 | .social_area{ position:absolute; right:0; top:0; background:#fff; border-top:1px solid #ccc; border-bottom:1px solid #ccc; border-right:1px solid #ccc; height:31px}
 86 | .social_nav{ text-align:right}
 87 | .social_nav li{ display:block; float:left}
 88 | .social_nav li a{ display:block; float:left; height:30px; text-indent:-9999px; width:30px; border-left:1px solid #ccc; -webkit-transition:all 0.5s; -moz-transition:all 0.5s; -ms-transition:all 0.5s;  -o-transition:all 0.5s; transition:all 0.5s}
 89 | .social_nav li.facebook a{ background-image:url("images/socials.png"); background-position:0 -30px; background-size:270px 60px}
 90 | .social_nav li.facebook a:hover{ background-position:0 0px; background-size:270px 60px; background-color:#436eac}
 91 | .social_nav li.twitter a{ background-image:url("images/socials.png"); background-position:-30px -30px; background-size:270px 60px}
 92 | .social_nav li.twitter a:hover{ background-position:-30px 0px; background-size:270px 60px; background-color:#0598c9}
 93 | .social_nav li.flickr a{ background-image:url("images/socials.png"); background-position:-60px -30px; background-size:270px 60px}
 94 | .social_nav li.flickr a:hover{ background-position:-60px 0px; background-size:270px 60px; background-color:#e33b7e}
 95 | .social_nav li.pinterest a{ background-image:url("images/socials.png"); background-position:-90px -30px; background-size:270px 60px}
 96 | .social_nav li.pinterest a:hover{ background-position:-90px 0px; background-size:270px 60px; background-color:#cb2027}
 97 | .social_nav li.googleplus a{ background-image:url("images/socials.png"); background-position:-120px -30px; background-size:270px 60px}
 98 | .social_nav li.googleplus a:hover{ background-position:-120px 0px; background-size:270px 60px; background-color:#d64b2e}
 99 | .social_nav li.vimeo a{ background-image:url("images/socials.png"); background-position:-150px -30px; background-size:270px 60px}
100 | .social_nav li.vimeo a:hover{ background-position:-150px 0px; background-size:270px 60px; background-color:#86ae24}
101 | .social_nav li.youtube a{ background-image:url("images/socials.png"); background-position:-180px -30px; background-size:270px 60px; width:60px}
102 | .social_nav li.youtube a:hover{ background-position:-180px 0px; background-size:270px 60px; background-color:#e32114}
103 | .social_nav li.mail a{ background-image:url("images/socials.png"); background-position:-240px -30px; background-size:270px 60px; width:32px}
104 | .social_nav li.mail a:hover{background-position:-240px 0px; background-size:270px 60px; background-color:#bc75d6}
105 | #sliderSection{background-color:#fff; display:inline; float:left; width:100%; padding:0 30px}
106 | .single_iteam{display:inline; float:left; position:relative; width:100%; height:448px}
107 | .single_iteam img{width:100%; height:100%}
108 | .single_iteam img:hover{opacity:1}
109 | .slider_article{ bottom:20px; left:0; position:absolute; right:0; padding:10px 15px}
110 | .slider_article > h2 a{ background:none repeat scroll 0 0 rgba(0,0,0,0.4); color:#fff; font-size:18px; padding:10px; display:inline-block}
111 | .slider_article > p{ background:none repeat scroll 0 0 rgba(0,0,0,0.4); color:#fff; padding:3px; display:inline-block}
112 | .slick-prev:before{ content:""}
113 | .slick-prev{ background-image:url(images/slider_prev.png); background-repeat:no-repeat; background-position:center; left:10px}
114 | .slick-next:before{ content:""}
115 | .slick-next{ background-image:url(images/slider_next.png); background-repeat:no-repeat; background-position:center; left:60px}
116 | .slick-prev, .slick-next{background-color:#000; top:10%; width:40px; height:40px}
117 | .slick-prev:hover, .slick-next:hover{opacity:0.5}
118 | .latest_post{float:left; display:inline; width:100%}
119 | .latest_post > h2{background:none repeat scroll 0 0 #151515; color:#fff; font-family:'Oswald',sans-serif; font-size:18px; margin-top:5px; font-weight:400; margin-bottom:10px; margin-left:0; padding:0; position:relative; text-align:center; text-transform:uppercase}
120 | .latest_post > h2 span{padding:4px 10px}
121 | .latest_postnav{height:auto !important; margin-top:20px}
122 | .latest_postnav li{margin-bottom:10px; float:left; width:100%}
123 | .latest_post_container{display:inline; float:left; height:430px; position:relative; width:100%}
124 | .latest_post_container:hover #prev-button, .latest_post_container:hover #next-button{display:block}
125 | #prev-button{cursor:pointer; font-size:20px; left:0; position:absolute; text-align:center; top:-10px; width:100%; display:none}
126 | #next-button{cursor:pointer; display:none; font-size:20px; left:0; position:absolute; text-align:center; bottom:0; width:100%}
127 | #contentSection{float:left; display:inline; width:100%; background-color:#fff; padding:0 30px}
128 | .left_content{float:left; display:inline; width:100%}
129 | .single_post_content{float:left; display:inline; width:100%; margin-bottom:20px}
130 | .single_post_content > h2{background:none repeat scroll 0 0 #151515; color:#fff; font-family:'Oswald',sans-serif; font-size:18px; font-weight:400; margin-bottom:10px; margin-left:0; margin-top:5px; padding:0; position:relative; text-align:center; text-transform:uppercase; margin-bottom:20px}
131 | .single_post_content > h2 span{padding:4px 10px}
132 | .single_post_content_left{float:left; display:inline; width:49%}
133 | .business_catgnav{}
134 | .business_catgnav li{float:left; display:block; width:100%}
135 | .bsbig_fig{width:100%}
136 | .bsbig_fig > a{display:block}
137 | .bsbig_fig > a img{width:100%}
138 | .bsbig_fig figcaption{color:#333; font-family:"Oswald",sans-serif; font-size:23px; font-weight:300; margin-top:10px; margin-bottom:10px}
139 | .single_post_content_right{float:right; display:inline; width:48%}
140 | .right_content{float:left; display:inline; width:100%; min-height:300px}
141 | .spost_nav{}
142 | .spost_nav li{float:left; display:block; width:100%; margin-bottom:10px}
143 | .spost_nav .media-left{width:100px; height:80px}
144 | .media-left > img{height:70px; width:90px}
145 | .spost_nav .media-body > a{font-family:"Oswald",sans-serif}
146 | .featured_img{position:relative}
147 | .overlay:hover{ background:none repeat scroll 0 0 rgba(0,0,0,0.4)}
148 | .overlay{ bottom:0; display:block; left:0; position:absolute; width:100%; z-index:2; height:100%; -webkit-transition:all 0.5s; -mz-transition:all 0.5s; -ms-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
149 | .fashion_technology_area{display:inline; float:left; width:100%}
150 | .fashion{float:left; display:inline; width:48%}
151 | .technology{float:right; display:inline; width:48%}
152 | .photo_grid{position:relative; margin:0 auto; max-width:1000px; list-style:none; text-align:center}
153 | .photograph_nav{margin-left:-11px}
154 | .photograph_nav li{display:block; float:left; margin-left:9px; width:32%}
155 | .photo_grid figure{position:relative; float:left; overflow:hidden; margin:10px 1%; margin-top:0; height:150px; width:100%; background:#3085a3; text-align:center; cursor:pointer}
156 | .photo_grid figure img{position:relative; display:block; min-height:100%; max-width:100%; width:100%; height:100% opacity:0.8}
157 | .photo_grid figure figcaption{padding:2em; color:#fff; text-transform:uppercase; font-size:1.25em; -webkit-backface-visibility:hidden; backface-visibility:hidden}
158 | .photo_grid figure figcaption::before, .photo_grid figure figcaption::after{pointer-events:none}
159 | .photo_grid figure figcaption{position:absolute; bottom:0; left:0; width:100%; height:100%}
160 | .photo_grid figure h2{word-spacing:-0.15em; font-weight:300}
161 | .photo_grid figure h2 span{font-weight:800}
162 | .photo_grid figure h2, .photo_grid figure a{margin:0}
163 | .photo_grid figcaption a{color:#fff; font-size:68.5%; letter-spacing:1px; display:block; margin-top:7px}
164 | figure.effect-layla img{height:390px}
165 | figure.effect-layla figcaption{padding:1.5em}
166 | figure.effect-layla figcaption::before, figure.effect-layla figcaption::after{position:absolute; content:''; opacity:0}
167 | figure.effect-layla figcaption::before{top:20px; right:15px; bottom:20px; left:15px; border-top:1px solid #fff; border-bottom:1px solid #fff; -webkit-transform:scale(0,1); transform:scale(0,1); -webkit-transform-origin:0 0; transform-origin:0 0}
168 | figure.effect-layla figcaption::after{top:9px; right:25px; bottom:9px; left:25px; border-right:1px solid #fff; border-left:1px solid #fff; -webkit-transform:scale(1,0); transform:scale(1,0); -webkit-transform-origin:100% 0; transform-origin:100% 0}
169 | figure.effect-layla h2{font-size:18px; padding-top:33%; -webkit-transition:-webkit-transform 0.35s; transition:transform 0.35s}
170 | figure.effect-layla a{ text-transform:none; -webkit-transform:translate3d(0,-10px,0); transform:translate3d(0,-10px,0)}
171 | figure.effect-layla img, figure.effect-layla h2{-webkit-transform:translate3d(0,-30px,0); transform:translate3d(0,-30px,0)}
172 | figure.effect-layla img, figure.effect-layla figcaption::before, figure.effect-layla figcaption::after, figure.effect-layla a{-webkit-transition:opacity 0.35s,-webkit-transform 0.35s; transition:opacity 0.35s,transform 0.35s}
173 | figure.effect-layla:hover img{opacity:0.7; -webkit-transform:translate3d(0,0,0); transform:translate3d(0,0,0)}
174 | figure.effect-layla:hover figcaption::before, figure.effect-layla:hover figcaption::after{opacity:1; -webkit-transform:scale(1); transform:scale(1)}
175 | figure.effect-layla:hover h2{padding-top:26%}
176 | figure.effect-layla:hover h2, figure.effect-layla:hover a{opacity:1; -webkit-transform:translate3d(0,-35px,0); transform:translate3d(0,-35px,0)}
177 | figure.effect-layla:hover figcaption::after, figure.effect-layla:hover h2, figure.effect-layla:hover a, figure.effect-layla:hover img{-webkit-transition-delay:0.15s; transition-delay:0.15s}
178 | .single_sidebar{float:left; display:inline; width:100%; margin-bottom:20px}
179 | .single_sidebar > h2{background:none repeat scroll 0 0 #151515; color:#fff; font-family:"Oswald",sans-serif; font-size:18px; font-weight:400; margin-bottom:10px; margin-left:0; margin-top:5px; padding:0; position:relative; text-align:center; text-transform:uppercase}
180 | .single_sidebar > h2 span{padding:4px 10px}
181 | .cat-item a{background:none repeat scroll 0 0 #e4e4e4; color:#888; display:block; float:left; border-bottom:none !important; font-size:13px; line-height:12px; margin:0 2px 2px 0; padding:12px 17px; -webkit-transition:all 0.5s; -mz-transition:all 0.5s; -ms-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
182 | .cat-item a:hover{color:#fff; text-decoration:none !important}
183 | .tab-content{margin-top:10px}
184 | .nav-tabs{background:none repeat scroll 0 0 #333; border-bottom:none}
185 | .nav-tabs > li{margin-bottom:0}
186 | .nav-tabs > li > a{margin-right:0; color:#fff; font-size:15px; border-radius:0; border:none; font-family:"Oswald",sans-serif; -webkit-transition:all 0.5s; -mz-transition:all 0.5s; -ms-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
187 | .nav-tabs > li > a:hover, .nav-tabs > li > a:focus{color:#fff !important; border:none}
188 | .nav-tabs > li.active > a, .nav-tabs > li.active > a:hover, .nav-tabs > li.active > a:focus{border-width:0; border-color:#ddd #ddd transparent; color:#fff}
189 | .vide_area{float:left; display:inline; width:100%}
190 | .sideAdd{display:block; float:left; height:250px; width:100%; margin-top:10px}
191 | .sideAdd > img{width:100%; height:100%}
192 | .single_sidebar ul li a{border-bottom:1px solid #333; display:block}
193 | .single_sidebar .spost_nav li a{border-bottom:none; float:left}
194 | #footer{display:inline; float:left; width:100%; margin-bottom:20px}
195 | .footer_top{background-color:#252525; color:#ddd; display:inline; float:left; padding:10px 30px 48px; width:100%}
196 | .footer_widget{display:inline; float:left; width:100%; min-height:310px}
197 | .footer_widget > h2{border-bottom:3px solid #666; font-family:Oswald,arial,Georgia,serif; font-size:16px; padding:10px 0; text-transform:uppercase}
198 | .tag_nav{}
199 | .tag_nav li{}
200 | .tag_nav li a{border-bottom:1px solid #ddd; color:#ccc; display:block; padding:6px 6px 6px 0; -webkit-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
201 | .tag_nav li a:hover{padding-left:10px}
202 | .footer_bottom{float:left; display:inline; width:100%; padding:10px 30px; background-color:#303030; color:#ccc}
203 | .copyright{float:left; width:50%; padding-top:5px; text-align:left}
204 | .copyright a{color:#ccc}
205 | .developer{float:right; width:50%; text-align:right; padding-top:5px; color:#ccc}
206 | .developer a{color:#ccc}
207 | .catgArchive{border:medium none; color:#fff; display:inline; float:left; font-weight:bold; padding:10px 15px; width:100%; margin-top:15px}
208 | .catgArchive option{background-color:#fff; font-weight:normal; padding:5px; color:#d083cf}
209 | .nav-tabs > li{display:inline-block; float:none; width:32.55%}
210 | .nav-tabs{ text-align:center}
211 | .pagination > li > a, .pagination > li > span{background-color:#000; border:1px solid #000; color:#eee; margin-right:5px; -webkit-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
212 | .pagination > li > a:hover, .pagination > li > a:focus{background-color:#d083cf; color:#fff; border-color:#d083cf}
213 | .pagination > li:first-child > a, .pagination > li:first-child > span{border-bottom-left-radius:0; border-top-left-radius:0}
214 | .pagination > li:last-child > a, .pagination > li:last-child > span{border-bottom-right-radius:0; border-top-right-radius:0}
215 | .single_page{float:left; display:inline; width:100%}
216 | .single_page > h1{color:#333; font-family:Oswald,arial,Georgia,serif; font-size:30px; line-height:1.4em; margin:10px 0 -10px; padding:0 0 4px; text-transform:uppercase}
217 | .post_commentbox{display:inline; float:left; width:100%; border-top:1px solid #ddd; border-bottom:1px solid #ddd; margin-top:20px; padding:5px 0px}
218 | .post_commentbox a, .post_commentbox span{color:#798992; font-size:11px; margin-right:5px}
219 | .post_commentbox a > i, .post_commentbox span > i{margin-right:5px}
220 | .breadcrumb{background-color:#303030; border-radius:0}
221 | .breadcrumb li a{color:#fff}
222 | .single_page_content{display:inline; float:left; padding-top:20px; width:100%; border-bottom:1px solid #ddd; padding-bottom:20px}
223 | .single_page_content > img{max-width:100%; width:320px; height:213px; margin-bottom:15px}
224 | .single_page_content ul{position:relative; padding-left:25px}
225 | .single_page_content ul li{line-height:25px}
226 | .single_page_content ul li:before{content:""; height:5px; left:5px; position:absolute; width:9px; margin-top:8px}
227 | .single_page_content ul li:hover{opacity:0.75}
228 | .single_page_content h2{line-height:35px}
229 | .single_page_content h3{line-height:30px}
230 | .single_page_content h4{line-height:25px}
231 | .single_page_content h4{line-height:20px}
232 | .social_link{display:inline; float:left; margin-bottom:25px; margin-top:20px; width:100%}
233 | .sociallink_nav{text-align:center}
234 | .sociallink_nav li{display:inline-block}
235 | .sociallink_nav li a{color:#fff; display:inline-block; font-size:17px; padding:8px 12px; margin:0 3px; -webkit-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
236 | .sociallink_nav li a:hover{-webkit-transform:rotate(50deg); -moz-transform:rotate(50deg); -ms-transform:rotate(50deg); -os-transform:rotate(50deg); transform:rotate(50deg)}
237 | .sociallink_nav li a:hover i{-webkit-transform:rotate(-50deg); -moz-transform:rotate(-50deg); -ms-transform:rotate(-50deg); -os-transform:rotate(-50deg); transform:rotate(-50deg)}
238 | .sociallink_nav li:nth-child(1) > a{background:none repeat scroll 0 0 #3b5998; padding:8px 15px}
239 | .sociallink_nav li:nth-child(2) > a{background:none repeat scroll 0 0 #00acee}
240 | .sociallink_nav li:nth-child(3) > a{background:none repeat scroll 0 0 #dd4b39}
241 | .sociallink_nav li:nth-child(4) > a{background:none repeat scroll 0 0 #0e76a8}
242 | .sociallink_nav li:nth-child(5) > a{background:none repeat scroll 0 0 #c92228}
243 | .related_post{display:inline; float:left; margin-top:0; width:100%; margin-bottom:20px}
244 | .related_post > h2{border-bottom:1px solid #e3e3e3; padding-bottom:5px}
245 | .related_post > h2 i{font-size:25px}
246 | .related_post .spost_nav li{width:32%; margin-right:10px}
247 | .related_post .spost_nav li:last-child{margin-right:0}
248 | .nav-slit a{position:fixed; top:50%; display:block; outline:none; text-align:left; z-index:1000; -webkit-transform:translateY(-50%); transform:translateY(-50%)}
249 | .nav-slit a.prev{left:0}
250 | .nav-slit a.next{right:0}
251 | .nav-slit .icon-wrap{position:relative; display:block; padding:45px 5px}
252 | .nav-slit .icon-wrap i{color:#fff; font-size:40px; display:inline-block}
253 | .nav-slit div{position:absolute; top:0; width:200px; height:100%; background-color:#939a9f; -webkit-transition:-webkit-transform 0.3s 0.3s; transition:transform 0.3s 0.3s; -webkit-perspective:1000px; perspective:1000px}
254 | .nav-slit a.prev div{left:0; -webkit-transform:translateX(-100%); transform:translateX(-100%)}
255 | .nav-slit a.next div{right:0; text-align:right; -webkit-transform:translateX(100%); transform:translateX(100%)}
256 | .nav-slit h3{position:absolute; top:100%; margin:0; padding:0 20px; height:30%; color:#fff; text-transform:uppercase; letter-spacing:1px; font-weight:500; font-size:0.75em; line-height:2.75; width:200px; text-align:left; overflow:hidden; padding-top:4px; -webkit-transition:-webkit-transform 0.3s; transition:transform 0.3s; -webkit-transform:rotateX(-90deg); transform:rotateX(-90deg); -webkit-transform-origin:50% 0; transform-origin:50% 0; -webki-backface-visibility:hidden; -webkit-backface-visibility:hidden; backface-visibility:hidden}
257 | .nav-slit img{left:0; position:absolute; top:0; width:100%; width:200px; height:130px}
258 | .nav-slit a:hover div{-webkit-transform:translateX(0); transform:translateX(0)}
259 | .nav-slit a:hover h3{-webkit-transition-delay:0.6s; transition-delay:0.6s; -webkit-transform:rotateX(0deg); transform:rotateX(0deg)}
260 | .error_page{float:left; display:inline; width:100%; text-align:center}
261 | .error_page > h3{text-transform:uppercase}
262 | .error_page > h1{font-size:110px}
263 | .error_page > p{font-size:15px; margin:0 auto; width:80%; margin-bottom:40px}
264 | .error_page > span{display:inline-block; height:2px; text-align:center; width:100px}
265 | .error_page > a{color:#fff; display:inline-block; padding:5px 10px}
266 | .error_page > a:hover{opacity:0.75}
267 | .contact_area{float:left; display:inline; width:100%; margin-bottom:30px}
268 | .contact_area > h2{color:#fff; display:inline-block; font-size:20px; padding:7px 10px 5px; text-transform:uppercase; margin-bottom:30px}
269 | .contact_area > p{margin-bottom:20px}
270 | .contact_form input[type="text"], .contact_form input[type="email"], .contact_form textarea{border-radius:0; margin-bottom:30px}
271 | .contact_form input[type="submit"]{border:medium none; color:#fff; height:35px; padding:5px 10px; -webkit-transition:all 0.5s; -o-transition:all 0.5s; transition:all 0.5s}
272 | .contact_form input[type="submit"]:hover{border:1px solid #d083cf}
273 | @media(max-width:1199px ){.logo_area{width:34%}.add_banner{width:580px}.nav-tabs > li{width:32.3%}.photograph_nav li{width:31.5%}
274 | @media(max-width:991px ){.add_banner{display:none}.logo_area{width:100%}.nav > li > a{padding:8px 8px}.latest_post_container{height:380px;overflow:hidden}#next-button{bottom:-2px}.single_iteam{height:415px}.photograph_nav li{width:47.7%}.related_post .spost_nav li{width:100%}.nav-tabs > li{width:31.9%}.nav-tabs > li > a{font-size:13px;padding-left:0 !important;padding-right:0 !important;text-align:center}}
275 | @media(max-width:767px ){.navbar-collapse{padding-left:15px}.mobile-show{display:block}.desktop-home{display:none}.navbar-inverse .navbar-nav > li > a{display:block}.header_top_left{width:100%}.header_top_right > p{display:none}.social_area{display:none}.single_iteam a{height:100%}.single_iteam a > img{height:100%}.error_page > a{margin-bottom:25px}.nav-tabs > li{width:32.6%}}
276 | @media(max-width:480px ){.top_nav{text-align:center}.single_post_content_left{width:100%}.single_post_content_right{width:100%}.fashion{width:100%}.technology{width:100%}.copyright{text-align:center;width:100%}.developer{text-align:center;width:100%}.single_iteam{height:300px}.photo_grid figure{height:200px}.photograph_nav li{width:100%;margin-left:0}.nav > li > a{padding:8px 12px}.nav-tabs > li{width:32.6%}}
277 | @media(max-width:360px ){.latest_newsarea span{font-size:12px;line-height:2.2em;padding:2px 10px 1px 10px}.single_iteam{height:210px}.slider_article > p{display:none}.error_page > span{width:80px}.nav-tabs > li{width:32.3%}.pagination > li > a,.pagination > li > span{padding:4px 8px}}
278 | @media(max-width:320px ){.sociallink_nav li a{padding:5px 10px}.sociallink_nav li:nth-child(1) > a{padding:5px 13px}.nav-tabs > li{width:32.1%}}


--------------------------------------------------------------------------------
/theme.css:
--------------------------------------------------------------------------------
 1 | .scrollToTop{background-color:#d083cf; color:#fff}
 2 | .scrollToTop:hover, .scrollToTop:focus{background-color:#fff; color:#d083cf; border-color:1px solid #d083cf}
 3 | .logo > span{color:#d083cf}
 4 | .navbar-inverse .navbar-nav > li > a:hover, .navbar-inverse .navbar-nav > li > a:focus{background-color:#d083cf}
 5 | .navbar-inverse .navbar-nav > .active > a, .navbar-inverse .navbar-nav > .active > a:hover, .navbar-inverse .navbar-nav > .active > a:focus{background-color:#d083cf}
 6 | .dropdown-menu > li > a:hover, .dropdown-menu > li > a:focus{background-color:#d083cf}
 7 | .navbar-inverse .navbar-nav > .open > a, .navbar-inverse .navbar-nav > .open > a:hover, .navbar-inverse .navbar-nav > .open > a:focus{background-color:#d083cf}
 8 | .dropdown-menu > li > a:hover, .dropdown-menu > li > a:focus{background-color:#d083cf}
 9 | .navbar-inverse .navbar-nav > .open > a, .navbar-inverse .navbar-nav > .open > a:hover, .navbar-inverse .navbar-nav > .open > a:focus{background-color:#d083cf}
10 | .latest_newsarea span{background:none repeat scroll 0 0 #d083cf}
11 | .latest_post > h2 span{background:none repeat scroll 0 0 #d083cf}
12 | #prev-button{color:#d083cf}
13 | #next-button{color:#d083cf}
14 | .single_post_content > h2 span{background:none repeat scroll 0 0 #d083cf}
15 | .single_sidebar > h2 span{ background:none repeat scroll 0 0 #d083cf}
16 | .bsbig_fig figcaption a:hover{color:#d083cf}
17 | .spost_nav .media-body > a:hover{color:#d083cf}
18 | .cat-item a:hover{background-color:#d083cf}
19 | .nav-tabs > li > a:hover, .nav-tabs > li > a:focus{background-color:#d083cf}
20 | .nav-tabs > li.active > a, .nav-tabs > li.active > a:hover, .nav-tabs > li.active > a:focus{background-color:#d083cf}
21 | .single_sidebar > ul > li a:hover{color:#d083cf}
22 | .footer_widget > h2:hover{color:#d083cf}
23 | .tag_nav li a:hover{color:#d083cf; border-color:#d083cf}
24 | .copyright a:hover{color:#d083cf}
25 | .post_commentbox a:hover, .post_commentbox span:hover{color:#d083cf}
26 | .breadcrumb{border-left:10px solid #d083cf}
27 | .single_page_content ul li:before{background-color:#d083cf}
28 | .single_page_content h2, .single_page_content h3, .single_page_content h4, .single_page_content h5, .single_page_content h6{color:#d083cf}
29 | .nav-slit .icon-wrap{background-color:#d083cf}
30 | .nav-slit h3{background:#d083cf}
31 | .catgArchive{background-color:#d083cf}
32 | .error_page > h3{color:#d083cf}
33 | .error_page > span{background:none repeat scroll 0 0 #d083cf}
34 | .error_page > a{background-color:#d083cf}
35 | .contact_area > h2{background-color:#d083cf}
36 | .contact_form input[type="submit"]{background-color:#d083cf}
37 | .contact_form input[type="submit"]:hover{background-color:#fff; color:#d083cf}
38 | .related_post > h2 i{color:#d083cf}
39 | .form-control:focus{border-color:#d083cf;  box-shadow:0 0px 1px #d083cf inset,0 0 5px #d083cf}


--------------------------------------------------------------------------------