├── doc2vec ├── models │ └── README.md ├── process_doc2vec_similarity.py └── README.md ├── use ├── models │ └── README.md ├── process_use_similarity.py └── README.md ├── .gitignore ├── tf-idf ├── process_tfidf_similarity.py └── README.md ├── bert ├── process_bert_similarity.py └── README.md ├── jaccard ├── process_jaccard_similarity.py └── README.md └── README.md /doc2vec/models/README.md: -------------------------------------------------------------------------------- 1 | 2 | Download Doc2vec pretrained model from (https://github.com/jhlau/doc2vec)[https://github.com/jhlau/doc2vec]. It can be either enwiki or apnews. -------------------------------------------------------------------------------- /use/models/README.md: -------------------------------------------------------------------------------- 1 | 2 | Donwload the pretrained model into this directory from (https://tfhub.dev/google/universal-sentence-encoder/4)[https://tfhub.dev/google/universal-sentence-encoder/4]. 3 | -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- 1 | # Compiled source # 2 | ################### 3 | *.com 4 | *.class 5 | *.dll 6 | *.exe 7 | *.o 8 | *.so 9 | 10 | # Packages # 11 | ############ 12 | # it's better to unpack these files and commit the raw source 13 | # git has its own built in compression methods 14 | *.7z 15 | *.dmg 16 | *.gz 17 | *.iso 18 | *.jar 19 | *.rar 20 | *.tar 21 | *.zip 22 | 23 | # Logs and databases # 24 | ###################### 25 | *.log 26 | *.sql 27 | *.sqlite 28 | 29 | # OS generated files # 30 | ###################### 31 | .DS_Store 32 | .DS_Store? 33 | ._* 34 | .Spotlight-V100 35 | .Trashes 36 | ehthumbs.db 37 | Thumbs.db -------------------------------------------------------------------------------- /use/process_use_similarity.py: -------------------------------------------------------------------------------- 1 | from sklearn.metrics.pairwise import cosine_similarity 2 | 3 | import tensorflow as tf 4 | import tensorflow_hub as hub 5 | 6 | base_document = "This is an example sentence for the document to be compared" 7 | documents = ["This is the collection of documents to be compared against the base_document"] 8 | 9 | def process_use_similarity(): 10 | filename = "./models/universal-sentence-encoder_4" 11 | 12 | model = hub.load(filename) 13 | 14 | base_embeddings = model([base_document]) 15 | 16 | embeddings = model(documents) 17 | 18 | scores = cosine_similarity(base_embeddings, embeddings).flatten() 19 | 20 | highest_score = 0 21 | highest_score_index = 0 22 | for i, score in enumerate(scores): 23 | if highest_score < score: 24 | highest_score = score 25 | highest_score_index = i 26 | 27 | most_similar_document = documents[highest_score_index] 28 | print("Most similar document by USE with the score:", most_similar_document, highest_score) 29 | 30 | process_use_similarity() 31 | -------------------------------------------------------------------------------- /tf-idf/process_tfidf_similarity.py: -------------------------------------------------------------------------------- 1 | from sklearn.feature_extraction.text import TfidfVectorizer 2 | from sklearn.metrics.pairwise import cosine_similarity 3 | 4 | base_document = "This is an example sentence for the document to be compared" 5 | documents = ["This is the collection of documents to be compared against the base_document"] 6 | 7 | def process_tfidf_similarity(): 8 | vectorizer = TfidfVectorizer() 9 | 10 | # To make uniformed vectors, both documents need to be combined first. 11 | documents.insert(0, base_document) 12 | embeddings = vectorizer.fit_transform(documents) 13 | 14 | cosine_similarities = cosine_similarity(embeddings[0:1], embeddings[1:]).flatten() 15 | 16 | highest_score = 0 17 | highest_score_index = 0 18 | for i, score in enumerate(cosine_similarities): 19 | if highest_score < score: 20 | highest_score = score 21 | highest_score_index = i 22 | 23 | 24 | most_similar_document = documents[highest_score_index] 25 | 26 | print("Most similar document by TF-IDF with the score:", most_similar_document, highest_score) 27 | 28 | process_tfidf_similarity() 29 | -------------------------------------------------------------------------------- /bert/process_bert_similarity.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | from sklearn.metrics.pairwise import cosine_similarity 3 | from nltk import sent_tokenize 4 | 5 | from sentence_transformers import SentenceTransformer 6 | 7 | base_document = "This is an example sentence for the document to be compared" 8 | documents = ["This is the collection of documents to be compared against the base_document"] 9 | 10 | def process_bert_similarity(): 11 | # This will download and load the pretrained model offered by UKPLab. 12 | model = SentenceTransformer('bert-base-nli-mean-tokens') 13 | 14 | # Although it is not explicitly stated in the official document of sentence transformer, the original BERT is meant for a shorter sentence. We will feed the model by sentences instead of the whole documents. 15 | sentences = sent_tokenize(base_document) 16 | base_embeddings_sentences = model.encode(sentences) 17 | base_embeddings = np.mean(np.array(base_embeddings_sentences), axis=0) 18 | 19 | vectors = [] 20 | for i, document in enumerate(documents): 21 | 22 | sentences = sent_tokenize(document) 23 | embeddings_sentences = model.encode(sentences) 24 | embeddings = np.mean(np.array(embeddings_sentences), axis=0) 25 | 26 | vectors.append(embeddings) 27 | 28 | print("making vector at index:", i) 29 | 30 | scores = cosine_similarity([base_embeddings], vectors).flatten() 31 | 32 | highest_score = 0 33 | highest_score_index = 0 34 | for i, score in enumerate(scores): 35 | if highest_score < score: 36 | highest_score = score 37 | highest_score_index = i 38 | 39 | most_similar_document = documents[highest_score_index] 40 | print("Most similar document by BERT with the score:", most_similar_document, highest_score) 41 | 42 | process_bert_similarity() 43 | -------------------------------------------------------------------------------- /jaccard/process_jaccard_similarity.py: -------------------------------------------------------------------------------- 1 | import string 2 | import nltk 3 | 4 | nltk.download('stopwords') 5 | nltk.download('wordnet') 6 | nltk.download('punkt') 7 | 8 | from nltk.corpus import stopwords 9 | from nltk.tokenize import word_tokenize 10 | from nltk.stem import WordNetLemmatizer 11 | 12 | lemmatizer = WordNetLemmatizer() 13 | 14 | base_document = "This is an example sentence for the document to be compared" 15 | documents = ["This is the collection of documents to be compared against the base_document"] 16 | 17 | def preprocess(text): 18 | # Steps: 19 | # 1. lowercase 20 | # 2. Lammetize. (It does not stem. Try to preserve structure not to overwrap with potential acronym). 21 | # 3. Remove stop words. 22 | # 4. Remove punctuations. 23 | # 5. Remove character with the length size of 1. 24 | 25 | lowered = str.lower(text) 26 | 27 | stop_words = set(stopwords.words('english')) 28 | word_tokens = word_tokenize(lowered) 29 | 30 | words = [] 31 | for w in word_tokens: 32 | if w not in stop_words: 33 | if w not in string.punctuation: 34 | if len(w) > 1: 35 | lemmatized = lemmatizer.lemmatize(w) 36 | words.append(lemmatized) 37 | 38 | return words 39 | 40 | def calculate_jaccard(word_tokens1, word_tokens2): 41 | # Combine both tokens to find union. 42 | both_tokens = word_tokens1 + word_tokens2 43 | union = set(both_tokens) 44 | 45 | # Calculate intersection. 46 | intersection = set() 47 | for w in word_tokens1: 48 | if w in word_tokens2: 49 | intersection.add(w) 50 | 51 | jaccard_score = len(intersection)/len(union) 52 | return jaccard_score 53 | 54 | def process_jaccard_similarity(): 55 | 56 | # Tokenize the base document we are comparing against. 57 | base_tokens = preprocess(base_document) 58 | 59 | # Tokenize each document 60 | all_tokens = [] 61 | for i, document in enumerate(documents): 62 | tokens = preprocess(document) 63 | all_tokens.append(tokens) 64 | 65 | print("making word tokens at index:", i) 66 | 67 | all_scores = [] 68 | for tokens in all_tokens: 69 | score = calculate_jaccard(base_tokens, tokens) 70 | 71 | all_scores.append(score) 72 | 73 | highest_score = 0 74 | highest_score_index = 0 75 | for i, score in enumerate(all_scores): 76 | if highest_score < score: 77 | highest_score = score 78 | highest_score_index = i 79 | 80 | most_similar_document = documents[highest_score_index] 81 | 82 | print("Most similar document by Jaccard with the score:", most_similar_document, highest_score) 83 | 84 | process_jaccard_similarity() 85 | -------------------------------------------------------------------------------- /doc2vec/process_doc2vec_similarity.py: -------------------------------------------------------------------------------- 1 | from gensim.models.doc2vec import Doc2Vec 2 | from sklearn.metrics.pairwise import cosine_similarity 3 | 4 | import string 5 | import nltk 6 | 7 | nltk.download('stopwords') 8 | nltk.download('wordnet') 9 | nltk.download('punkt') 10 | 11 | from nltk.corpus import stopwords 12 | from nltk.tokenize import word_tokenize 13 | from nltk.stem import WordNetLemmatizer 14 | 15 | lemmatizer = WordNetLemmatizer() 16 | 17 | base_document = "This is an example sentence for the document to be compared" 18 | documents = ["This is the collection of documents to be compared against the base_document"] 19 | 20 | def preprocess(text): 21 | # Steps: 22 | # 1. lowercase 23 | # 2. Lammetize. (It does not stem. Try to preserve structure not to overwrap with potential acronym). 24 | # 3. Remove stop words. 25 | # 4. Remove punctuations. 26 | # 5. Remove character with the length size of 1. 27 | 28 | lowered = str.lower(text) 29 | 30 | stop_words = set(stopwords.words('english')) 31 | word_tokens = word_tokenize(lowered) 32 | 33 | words = [] 34 | for w in word_tokens: 35 | if w not in stop_words: 36 | if w not in string.punctuation: 37 | if len(w) > 1: 38 | lemmatized = lemmatizer.lemmatize(w) 39 | words.append(lemmatized) 40 | 41 | return words 42 | 43 | def process_doc2vec_similarity(): 44 | 45 | # Both pretrained models are publicly available at public repo of jhlau. 46 | # URL: https://github.com/jhlau/doc2vec 47 | 48 | # filename = './models/apnews_dbow/doc2vec.bin' 49 | filename = './models/enwiki_dbow/doc2vec.bin' 50 | 51 | model= Doc2Vec.load(filename) 52 | 53 | tokens = preprocess(base_document) 54 | 55 | # Only handle words that appear in the doc2vec pretrained vectors. enwiki_ebow model contains 669549 vocabulary size. 56 | tokens = list(filter(lambda x: x in model.wv.vocab.keys(), tokens)) 57 | 58 | base_vector = model.infer_vector(tokens) 59 | 60 | vectors = [] 61 | for i, document in enumerate(documents): 62 | 63 | tokens = preprocess(document) 64 | tokens = list(filter(lambda x: x in model.wv.vocab.keys(), tokens)) 65 | vector = model.infer_vector(tokens) 66 | vectors.append(vector) 67 | 68 | print("making vector at index:", i) 69 | 70 | scores = cosine_similarity([base_vector], vectors).flatten() 71 | 72 | highest_score = 0 73 | highest_score_index = 0 74 | for i, score in enumerate(scores): 75 | if highest_score < score: 76 | highest_score = score 77 | highest_score_index = i 78 | 79 | most_similar_document = documents[highest_score_index] 80 | print("Most similar document by Doc2vec with the score:", most_similar_document, highest_score) 81 | 82 | process_doc2vec_similarity() 83 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Document Similarity Algorithms Experiment 2 | 3 | Document similarity comparison using 5 popular algorithms: Jaccard, TF-IDF, Doc2vec, USE, and BERT. 4 | 5 | 33,914 New York Times articles are used for the experiment. It aims to show which algorithm yields the best result out of the box in 2020. 6 | 7 | # Purpose 8 | 9 | 1. By running multiple algorithms with some of which are most up-to-date and trendy in NLP community, it will show which algorithms give the best results and by how much onto the same set of data. 10 | 2. By using full-length popular media publications as our data input, we will simulate the real world similarity/recommendation use case. 11 | 3. By following URLs, you can actually see and compare the quality document similarity yourself. 12 | 4. By using only the pretrained models available publicly, you can easily set it up and start implementing document similarity on your own with very little prior knowledge while expecting the similar result output. 13 | 14 | # Data Setup 15 | 16 | 33,914 New York Times articles from 2018 to June 2020 were selected. The data was mostly collected from their RSS feed. 17 | 18 | 5 articles were chosen as the basis. They represent different categories. 19 | 20 | 1. [How My Worst Date Ever Became My Best](https://www.nytimes.com/2020/02/14/style/modern-love-worst-date-of-my-life-became-best.html) (lifestyle) 21 | 2. [A Deep-Sea Magma Monster Gets a Body Scan](https://www.nytimes.com/2019/12/03/science/axial-volcano-mapping.html) (science) 22 | 3. [Renault and Nissan Try a New Way After Years When Carlos Ghosn Ruled](https://www.nytimes.com/2019/11/29/business/renault-nissan-mitsubishi-alliance.html) (business) 23 | 4. [Dominic Thiem Beats Rafael Nadal in Australian Open Quarterfinal](https://www.nytimes.com/2020/01/29/sports/tennis/thiem-nadal-australian-open.html) (sports) 24 | 5. [2020 Democrats Seek Voters in an Unusual Spot: Fox News](https://www.nytimes.com/2019/04/17/us/politics/fox-news-democrats-2020.html) (politics) 25 | 26 | # Comparison Criteria 27 | 28 | Overlapping tags, sections, subsections, writing format, and subjective judgement are considered. For a more detailed description, please follow this blog post. 29 | 30 | # Algorithm That Win Overall 31 | 32 | TF-IDF. It resulted in the best matches in 2.5 out of 5 comparisons. 33 | 34 | The detailed breakdowns of how each algorithm predicted can be found in the algorithm folders. 35 | 36 | # Winner Algorithm By Each Article 37 | 38 | ## How My Worst Date Ever Became My Best 39 | 40 | Winner: BERT 41 | 42 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 43 | |---|---|---|---|---|---|---| 44 | | [Why Are All the Exes Texting?](https://www.nytimes.com/2020/05/29/style/modern-love-coronavirus-why-are-all-the-exes-texting.html) | 1 | Y | Y | Y | Dating | Related | 45 | | [When Love Seems Too Easy to Trust](https://www.nytimes.com/2018/04/13/style/when-love-seems-too-easy-to-trust.html) | 1 | Y | Y | Y | Dating | Related | 46 | | [He Saved His Last Lesson for Me](https://www.nytimes.com/2020/03/06/style/modern-love-long-distance-dating-korea.html) | 1 | Y | Y | Y | Dating | Related | 47 | 48 | ## A Deep-Sea Magma Monster Gets a Body Scan 49 | 50 | Winner: TF-IDF 51 | 52 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 53 | |---|---|---|---|---|---|---| 54 | | [A 3D Encounter With a Violent Volcano’s Underbelly](https://www.nytimes.com/2019/12/18/science/volcano-3d-reunion-island.html) | 4 | Y | N | Y | 3D Mapped Volcano | Highly Related | 55 | | [Pressure, and Mystery, on the Rise](https://www.nytimes.com/2015/01/06/science/predicting-what-a-volcano-may-or-may-not-do-is-as-tricky-as-it-is-crucial-as-iceland-well-knows.html) | 1 | Y | Y | Y | Iceland's Volcano | Related | 56 | | [It’s Not Just Hawaii: The U.S. Has 169 Volcanoes That Could Erupt](https://www.nytimes.com/2018/05/14/us/us-active-volcanoes-hawaii.html) | 2 | N | N | Y | Volcanos | Related | 57 | 58 | 59 | ## Renault and Nissan Try a New Way After Years When Carlos Ghosn Ruled 60 | 61 | Winner: TF-IDF 62 | 63 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 64 | |---|---|---|---|---|---|---| 65 | | [Nissan CEO Says 'No Merit' in Merger With Renault-Nikkei](https://www.nytimes.com/reuters/2018/04/25/business/25reuters-renault-nissan-m-a.html) | 3 | N | N | Y | Nissan and Renault | Related | 66 | | [Carlos Ghosn and the Roots of Nissan’s Decline](https://www.nytimes.com/2020/01/15/automobiles/nissan-carlos-ghosn-strategy.html) | 3 | N | N | N | Carlos Ghosn | Related | 67 | | [Renault Chooses Volkswagen Executive as New C.E.O.](https://www.nytimes.com/2020/01/28/business/renault-ceo-luca-de-meo.html) | 5 | Y | Y | Y | Renault CEO | Very Related | 68 | 69 | ## Dominic Thiem Beats Rafael Nadal in Australian Open Quarterfinal 70 | 71 | Winner: Jaccard, TF-IDF, and USE 72 | 73 | (Jaccard) 74 | 75 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 76 | |---|---|---|---|---|---|---| 77 | | [Djokovic vs. Federer, a Rivalry for the Ages, Is One-Sided This Time](https://www.nytimes.com/2020/01/30/sports/tennis/djokovic-federer-australian-open.html) | 3 | Y | Y | Y | Australian Open | Related | 78 | | [Novak Djokovic Wins the Australian Open](https://www.nytimes.com/2020/02/02/sports/tennis/australian-open-djokovic-thiem.html) | 4 | Y | Y | Y | Dominic vs Novak in Australian Open | Very Related | 79 | | [With Rome Title, Nadal Back on Track Entering French Open](https://www.nytimes.com/aponline/2018/05/20/sports/tennis/ap-ten-italian-open.html) | 1 | N | N | Y | French Open | Unrelated | 80 | 81 | ## 2020 Democrats Seek Voters in an Unusual Spot: Fox News 82 | 83 | Winner: USE 84 | 85 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 86 | |---|---|---|---|---|---|---| 87 | | [Bernie Sanders Had a Problem With MSNBC. Then Came Super Tuesday.](https://www.nytimes.com/2020/03/05/business/media/msnbc-bernie-sanders-media.html) | 7 | N | N | Y | Sanders and MSNBC | Very Related | 88 | | [Democrats, Don’t Abandon Fox News](https://www.nytimes.com/2019/03/08/opinion/fox-news-democrats-debate.html) | 7 | N | N | N | Democrats and Fox News | Very Related | 89 | | [Candidates Running Against, and With, Cable News](https://www.nytimes.com/2010/10/24/us/politics/24cable.html) | 4 | Y | Y | Y | Fox, MSNBC and politics | Related | 90 | -------------------------------------------------------------------------------- /use/README.md: -------------------------------------------------------------------------------- 1 | # Experiment 2 | 3 | 5 articles were compared against 33,914 articles with Universal Sentence Encoder (USE) similarity. 4 | 5 | # [How My Worst Date Ever Became My Best](https://www.nytimes.com/2020/02/14/style/modern-love-worst-date-of-my-life-became-best.html) 6 | 7 | - Theme: Dating, divorced 8 | - Tags: Dating 9 | - Section: style 10 | - Subsection: modern love 11 | - Story: Human interest 12 | - best case prediction: The article will be about dating for older couples who have experienced marriage. 13 | 14 | ### Result 15 | 16 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 17 | |---|---|---|---|---|---|---| 18 | | [My Quarantine Boyfriend Lost Everything (but Found Me)](https://www.nytimes.com/2020/06/05/style/modern-love-coronavirus-quarantine-boyfriend-lost-everything.html) | 1 | Y | Y | Y | Dating | Related | 19 | | [Flying Close to Temptation](https://www.nytimes.com/2018/04/20/style/modern-love-flying-close-to-temptation.html) | 0 | Y | Y | Y | LGBTQ | Related | 20 | | [Why Are All the Exes Texting?](https://www.nytimes.com/2020/05/29/style/modern-love-coronavirus-why-are-all-the-exes-texting.html) | 1 | Y | Y | Y | Dating | Related | 21 | 22 | 1. My Quarantine Boyfriend Lost Everything (but Found Me) 23 | 24 | - Tas: Love,Dating,Coronavirus;COVID-19,NYC 25 | - Section: style 26 | - Subsection: modern love 27 | - Story: Human interest 28 | 29 | 2. Flying Close to Temptation 30 | 31 | - Tags: Single People,bars;nightclubs,Love,Alcohol,Alcohol abuse,Gay and Lesbian;LGBT;LGBTQ;Gay;Lesbian,MarriageGinsberg,Women and Girls 32 | - Section: style 33 | - Subsection: modern love 34 | - Story: human interest 35 | 36 | 3. Why Are All the Exes Texting? 37 | 38 | - Tags: Dating,Quarantines;Social Distancing;Shelter at home,Coronavirus;COVID-19 39 | - Section: style 40 | - Subsection: modern love 41 | - Story: human interest 42 | 43 | # [A Deep-Sea Magma Monster Gets a Body Scan](https://www.nytimes.com/2019/12/03/science/axial-volcano-mapping.html) 44 | 45 | - Theme: Nature, 3D scan 46 | - Tags: Axial Seamount,Volcano,Oceans and Seas,Geology,3-D,Ships and Shipping,Sensor,Scanners,Pacific Ocean,Oregon,your-feed-science 47 | - Section: science 48 | - Subsection: none 49 | - Story: Informational 50 | - best case prediction: The article will talk about ocean and 3d scanning. 51 | 52 | ### Result 53 | 54 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 55 | |---|---|---|---|---|---|---| 56 | | [A 3D Encounter With a Violent Volcano’s Underbelly](https://www.nytimes.com/2019/12/18/science/volcano-3d-reunion-island.html) | 4 | Y | N | Y | 3D Mapped Volcano | Highly Related | 57 | | [The Mount St. Helens Eruption Was the Volcanic Warning We Needed](https://www.nytimes.com/2020/05/18/science/mt-st-helens-eruption.html) | 3 | Y | Y | N | Volcano | Related | 58 | | [Kilauea Volcano’s Lava Fields Offer Scientists a Portal to Mars](https://www.nytimes.com/2018/05/28/science/kilauea-volcanos-lava-fields-scientists-mars.html) | 1 | Y | Y | Y | Volcano and Mars | Related | 59 | 60 | ### Breakdown 61 | 62 | 1. A 3D Encounter With a Violent Volcano’s Underbelly 63 | 64 | - Tags: Volcano,Reunion Island,3-D,Helicopter,Research,Geology,Scientific Reports Journal,Marc Dumont,your-feed-science 65 | - Section: science 66 | - Subsection: TRILOBITES 67 | - Story: Informational 68 | 69 | 2. The Mount St. Helens Eruption Was the Volcanic Warning We Needed 70 | 71 | - Tags: your-feed-science,Volcano,Geology,Mount St Helens,Washington 72 | - Section: science 73 | - Subsection: none 74 | - Story: Past events 75 | 76 | 3. Kilauea Volcano’s Lava Fields Offer Scientists a Portal to Mars 77 | 78 | - Tags: Space,Volcano,Mars,NASA 79 | - Section: science 80 | - Subsection: none 81 | - Story: informational 82 | 83 | # [Renault and Nissan Try a New Way After Years When Carlos Ghosn Ruled](https://www.nytimes.com/2019/11/29/business/renault-nissan-mitsubishi-alliance.html) 84 | 85 | - Theme: Renault, Nissan and Ghosn 86 | - Tags: Cars,Appointments and Executive Changes,Nissan Motor,Renault,Carlos Ghosn,Jean-Dominique Senard,Makoto Uchida,Mitsubishi Motor 87 | - Section: buisness 88 | - Subsection: none 89 | - Story: News 90 | - Best case prediction: The artile will talk about what has happened to either Renault and Nissan in relation to Carlos Ghosn incident. 91 | 92 | ### Result 93 | 94 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 95 | |---|---|---|---|---|---|---| 96 | | Renault Chooses Volkswagen Executive as New C.E.O. | 5 | Y | Y | Y | Renault CEO | Very Related | 97 | | Renault and Nissan Unveil Plan to Patch Tattered Alliance | 4 | Y | Y | Y | Nissan and Renault | Very Related | 98 | | Fiat Chrysler and Peugeot Agree on Terms to Forge New Auto Giant | 2 | Y | Y | Y | Car Industry | Unrelated | 99 | 100 | ### Breakdown 101 | 102 | 1. Renault Chooses Volkswagen Executive as New C.E.O. 103 | 104 | - Tags: Cars,Appointments and Executive Changes,Nissan Motor,Renault,Volkswagen,Jean-Dominique Senard,France,Luca de Meo 105 | - Section: business 106 | - Subsection: none 107 | - Story: News 108 | 109 | 2. Renault and Nissan Unveil Plan to Patch Tattered Alliance 110 | 111 | - Tags: Cars,Nissan Motor,Renault,Electric Cars and Hybrids,Carlos Ghosn,Jean-Dominique Senard,Coronavirus;COVID-19 112 | - Section: business 113 | - Subsection: none 114 | - Story: News 115 | 116 | 3. Fiat Chrysler and Peugeot Agree on Terms to Forge New Auto Giant 117 | 118 | - Tags: Cars,Mergers and Acquisitions,,GM,Peugeot,John Elkann,Carlos Tavares,France 119 | - Section: business 120 | - Subsection: none 121 | - Story: News 122 | - Note: Not related to Nissan. No mention of it. Only Renault and French government. 123 | 124 | # [Dominic Thiem Beats Rafael Nadal in Australian Open Quarterfinal](https://www.nytimes.com/2020/01/29/sports/tennis/thiem-nadal-australian-open.html) 125 | 126 | - Theme: Tennis, Dominic Thiem, Australian Open 127 | - Tags: Tennis,Australian Open,Novak Djokovic,Dominic Thiem 128 | - Section: sports 129 | - Subsection: tennis 130 | - Story: News 131 | - Best case prediction: The artile will talk about Dominic Thiem and Australian Open. 132 | 133 | ### Result 134 | 135 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 136 | |---|---|---|---|---|---|---| 137 | | [Djokovic vs. Federer, a Rivalry for the Ages, Is One-Sided This Time](https://www.nytimes.com/2020/01/30/sports/tennis/djokovic-federer-australian-open.html) | 3 | Y | Y | Y | Australian Open | Related | 138 | | [Novak Djokovic Wins the Australian Open](https://www.nytimes.com/2020/02/02/sports/tennis/australian-open-djokovic-thiem.html) | 4 | Y | Y | Y | Dominic vs Novak in Australian Open | Very Related | 139 | | [Stan Wawrinka and Rafael Nadal Reach French Open Final](https://www.nytimes.com/2017/06/09/sports/tennis/french-open-stan-wawrinka-semifinals.html) | 1 | Y | Y | Y | French Open | Unrelated | 140 | 141 | ### Breakdown 142 | 143 | 1. Djokovic vs. Federer, a Rivalry for the Ages, Is One-Sided This Time 144 | 145 | - Tags: Tennis,Australian Open,Novak Djokovic,Roger Federer 146 | - Section: sports 147 | - Subsection: tennis 148 | - Story: News 149 | - Note: Same on Australian Open. Not about Dominic 150 | 151 | 152 | 2. Novak Djokovic Wins the Australian Open 153 | 154 | - Tags: Novak Djokovic,Dominic Thiem,Australian Open,Tennis 155 | - Section: sports 156 | - Subsection: tennis 157 | - Story: News 158 | 159 | 3. Stan Wawrinka and Rafael Nadal Reach French Open Final 160 | 161 | - Tags: French Open,Stan Wawrinka,Andy Murray,Tennis 162 | - Section: sports 163 | - Subsection: tennis 164 | - Story: News 165 | - Note: from 2017. Mention about Dominic 166 | 167 | # [2020 Democrats Seek Voters in an Unusual Spot: Fox News](https://www.nytimes.com/2019/04/17/us/politics/fox-news-democrats-2020.html) 168 | 169 | - Theme: Bernie Sander, Democrat, Fox 170 | - Tags: Fox News Channel,Democrats,2020 Election,News media;journalism,TV,Bernard Sanders;Bernie Sanders,Presidential debates;Democratic debates;2020 debates 171 | - Section: US 172 | - Subsection: politics 173 | - Story: News 174 | - Best case prediction: The artile will talk about Bernie Sander, Democrats and Fox News in 2020 election. 175 | 176 | ### Result 177 | 178 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 179 | |---|---|---|---|---|---|---| 180 | | [Bernie Sanders Had a Problem With MSNBC. Then Came Super Tuesday.](https://www.nytimes.com/2020/03/05/business/media/msnbc-bernie-sanders-media.html) | 7 | N | N | Y | Sanders and MSNBC | Very Related | 181 | | [Democrats, Don’t Abandon Fox News](https://www.nytimes.com/2019/03/08/opinion/fox-news-democrats-debate.html) | 7 | N | N | N | Democrats and Fox News | Very Related | 182 | | [Candidates Running Against, and With, Cable News](https://www.nytimes.com/2010/10/24/us/politics/24cable.html) | 4 | Y | Y | Y | Fox, MSNBC and politics | Related | 183 | 184 | ### Breakdown 185 | 186 | 1. Bernie Sanders Had a Problem With MSNBC. Then Came Super Tuesday. 187 | 188 | - Tags: Bernard Sanders;Bernie Sanders,Democrats,2020 Election,MSNBC,News media;journalism,Primaries,TV,Rachel Maddow,Chris Matthews 189 | - Section: business 190 | - Subsection: media 191 | - Story: News 192 | - Note: Very good. Another right media appearance by Democrat 193 | 194 | 2. Democrats, Don’t Abandon Fox News 195 | 196 | - Tags: Democrats,Fox News Channel,Presidential debates;Democratic debates;2020 debates,2020 Election,News media;journalism,US Politics 197 | - Section: opinion 198 | - Subsection: none 199 | - Story: Informational 200 | 201 | 3. Candidates Running Against, and With, Cable News 202 | 203 | - Tags: Campaign advertising,US Politics,News Corp,MSNBC,Republican Governors Assn,Cable television,News media;journalism,Campaign finance,Fox News Channel,Election,Rupert Murdoch 204 | - Section: us 205 | - Subsection: politics 206 | - Story: News 207 | - Note: Old. Mention democrat. 208 | -------------------------------------------------------------------------------- /jaccard/README.md: -------------------------------------------------------------------------------- 1 | # Experiment 2 | 3 | 5 articles were compared against 33,914 articles with Jaccard similarity. 4 | 5 | # [How My Worst Date Ever Became My Best](https://www.nytimes.com/2020/02/14/style/modern-love-worst-date-of-my-life-became-best.html) 6 | 7 | - Theme: Dating, divorced 8 | - Tags: Dating 9 | - Section: style 10 | - Subsection: modern love 11 | - Story: Human interest 12 | - best case prediction: The article will be about dating for older couples who have experienced marriage. 13 | 14 | ### Result 15 | 16 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 17 | |---|---|---|---|---|---|---| 18 | | [Life Isn’t Like the Movies (Even if You Write the Movies)](https://www.nytimes.com/2020/02/28/style/modern-love-nancy-meyers-rom-com.html) | 1 | Y | Y | Y | Unhappy dating | Very Related | 19 | | [Telling the Truth Wasn’t an Option](https://www.nytimes.com/2020/03/20/style/modern-love-undocumented-immigrant-telling-truth-wasnt-option.html) | 1 | Y | Y | Y | Illegal Immigration | Related | 20 | | [Does It Matter How Many We Are?](https://www.nytimes.com/2019/11/29/style/modern-love-a-family-isnt-a-number.html) | 0 | Y | Y | Y | Family | Unrelated | 21 | 22 | ### Breakdown 23 | 24 | 1. Life Isn’t Like the Movies (Even if You Write the Movies) 25 | 26 | - Tags: Love,Dating,Weddings and Engagements,Movies,Divorce 27 | - Section: style 28 | - Subsection: modern love 29 | - Story: Human interest 30 | 31 | 2. Telling the Truth Wasn’t an Option 32 | 33 | - Tags: Love,Dating,Illegal Immigration;Undocumented Migrants 34 | - Section: style 35 | - Subsection: modern love 36 | - Story: Human interest 37 | 38 | 3. Does It Matter How Many We Are? 39 | 40 | - Tags: Love,Family,Thanksgiving,Grief 41 | - Section: style 42 | - Subsection: modern love 43 | - Story: Human interest 44 | - Note: The focus is on family and children rather than dating. 45 | 46 | # [A Deep-Sea Magma Monster Gets a Body Scan](https://www.nytimes.com/2019/12/03/science/axial-volcano-mapping.html) 47 | 48 | - Theme: Nature, 3D scan 49 | - Tags: Axial Seamount,Volcano,Oceans and Seas,Geology,3-D,Ships and Shipping,Sensor,Scanners,Pacific Ocean,Oregon,your-feed-science 50 | - Section: science 51 | - Subsection: none 52 | - Story: Informational 53 | - best case prediction: The article will talk about ocean and 3d scanning. 54 | 55 | ### Result 56 | 57 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 58 | |---|---|---|---|---|---|---| 59 | | [Pressure, and Mystery, on the Rise](https://www.nytimes.com/2015/01/06/science/predicting-what-a-volcano-may-or-may-not-do-is-as-tricky-as-it-is-crucial-as-iceland-well-knows.html) | 1 | Y | Y | Y | Iceland's Volcano | Related | 60 | | [What Could We Lose if a NASA Climate Mission Goes Dark?](https://www.nytimes.com/2017/09/12/magazine/what-could-we-lose-if-a-nasa-mission-goes-dark.html) | 0 | N | N | Y | Trump, NASA, Climate Change | Unrelated | 61 | | [A Deep-Diving Sub. A Deadly Fire. And Russia’s Secret Undersea Agenda.](https://www.nytimes.com/2020/04/20/world/europe/russian-submarine-fire-losharik.html) | 1 | N | N | Y | Military conflict | Unrelated | 62 | 63 | ### Breakdown 64 | 65 | 1. Pressure, and Mystery, on the Rise 66 | 67 | - Tags: Volcano,Iceland 68 | - Section: science 69 | - Subsection: none 70 | - Story: Informational 71 | 72 | 2. What Could We Lose if a NASA Climate Mission Goes Dark? 73 | 74 | - Tags: NASA,Climate Change,Global Warming,Satellite,Space,Federal Budget,Donald Trump,US Politics 75 | - Section: magazine 76 | - Subsection: feature 77 | - Story: Informational 78 | 79 | 3. A Deep-Diving Sub. A Deadly Fire. And Russia’s Secret Undersea Agenda. 80 | 81 | - Tags: Russia,Submarines and Submersibles,Boat Accidents,Military,Arctic,US Military,Norway,US 82 | - Section: world 83 | - Subsection: europe 84 | - Story: Informational 85 | 86 | # [Renault and Nissan Try a New Way After Years When Carlos Ghosn Ruled](https://www.nytimes.com/2019/11/29/business/renault-nissan-mitsubishi-alliance.html) 87 | 88 | - Theme: Renault, Nissan and Ghosn 89 | - Tags: Cars,Appointments and Executive Changes,Nissan Motor,Renault,Carlos Ghosn,Jean-Dominique Senard,Makoto Uchida,Mitsubishi Motor 90 | - Section: buisness 91 | - Subsection: none 92 | - Story: News 93 | - Best case prediction: The artile will talk about what has happened to either Renault and Nissan in relation to Carlos Ghosn incident. 94 | 95 | ### Result 96 | 97 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 98 | |---|---|---|---|---|---|---| 99 | | [Renault Chooses Volkswagen Executive as New C.E.O.](https://www.nytimes.com/2020/01/28/business/renault-ceo-luca-de-meo.html) | 5 | Y | Y | Y | Renault CEO | Very Related | 100 | | [Fiat Chrysler and Peugeot Agree on Terms to Forge New Auto Giant](https://www.nytimes.com/2019/12/18/business/fiat-chrysler-peugeot.html) | 2 | Y | Y | Y | Car Industry | Unrelated | 101 | | [Toyota and Suzuki Say They Are Considering an Alliance](https://www.nytimes.com/2016/10/13/business/international/toyota-suzuki-alliance-japan.html) | 1 | Y | N | Y | Car Industry | Unrelated | 102 | 103 | ### Breakdown 104 | 105 | 1. Renault Chooses Volkswagen Executive as New C.E.O. 106 | 107 | - Tags: Cars,Appointments and Executive Changes,Nissan Motor,Renault,Volkswagen,Jean-Dominique Senard,France,Luca de Meo 108 | - Section: business 109 | - Subsection: none 110 | - Story: News 111 | 112 | 2. Fiat Chrysler and Peugeot Agree on Terms to Forge New Auto Giant 113 | 114 | - Tags: Cars,Mergers and Acquisitions,,GM,Peugeot,John Elkann,Carlos Tavares,France 115 | - Section: business 116 | - Subsection: none 117 | - Story: News 118 | 119 | 3. Toyota and Suzuki Say They Are Considering an Alliance 120 | 121 | - Tags: Toyota Motor,Japan,Mergers and Acquisitions,Cars,Suzuki Motor 122 | - Section: business 123 | - Subsection: international 124 | - Story: News 125 | - Note: No mention about Nissan. Older. 126 | 127 | # [Dominic Thiem Beats Rafael Nadal in Australian Open Quarterfinal](https://www.nytimes.com/2020/01/29/sports/tennis/thiem-nadal-australian-open.html) 128 | 129 | - Theme: Tennis, Dominic Thiem, Australian Open 130 | - Tags: Tennis,Australian Open,Novak Djokovic,Dominic Thiem 131 | - Section: sports 132 | - Subsection: tennis 133 | - Story: News 134 | - Best case prediction: The artile will talk about Dominic Thiem and Australian Open. 135 | 136 | ### Result 137 | 138 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 139 | |---|---|---|---|---|---|---| 140 | | [Djokovic vs. Federer, a Rivalry for the Ages, Is One-Sided This Time](https://www.nytimes.com/2020/01/30/sports/tennis/djokovic-federer-australian-open.html) | 3 | Y | Y | Y | Australian Open | Related | 141 | | [Novak Djokovic Wins the Australian Open](https://www.nytimes.com/2020/02/02/sports/tennis/australian-open-djokovic-thiem.html) | 4 | Y | Y | Y | Dominic vs Novak in Australian Open | Very Related | 142 | | [With Rome Title, Nadal Back on Track Entering French Open](https://www.nytimes.com/aponline/2018/05/20/sports/tennis/ap-ten-italian-open.html) | 1 | N | N | Y | French Open | Unrelated | 143 | 144 | ### Breakdown 145 | 146 | 1. Djokovic vs. Federer, a Rivalry for the Ages, Is One-Sided This Time 147 | 148 | - Tags: Tennis,Australian Open,Novak Djokovic,Roger Federer 149 | - Section: sports 150 | - Subsection: tennis 151 | - Story: News 152 | - Note: Same on Australian Open. Not about Dominic 153 | 154 | 2. Novak Djokovic Wins the Australian Open 155 | 156 | - Tags: Novak Djokovic,Dominic Thiem,Australian Open,Tennis 157 | - Section: sports 158 | - Subsection: tennis 159 | - Story: News 160 | 161 | 3. With Rome Title, Nadal Back on Track Entering French Open 162 | 163 | - Tags: Paris,Madrid,Sports - Europe,Dominic Thiem,Rome,Rafael Nadal,Sports 164 | - Section: none 165 | - Subsection: none 166 | - Story: News 167 | - Note: From 2018. Mention Dominic Thiem. 168 | 169 | # [2020 Democrats Seek Voters in an Unusual Spot: Fox News](https://www.nytimes.com/2019/04/17/us/politics/fox-news-democrats-2020.html) 170 | 171 | - Theme: Bernie Sander, Democrat, Fox 172 | - Tags: Fox News Channel,Democrats,2020 Election,News media;journalism,TV,Bernard Sanders;Bernie Sanders,Presidential debates;Democratic debates;2020 debates 173 | - Section: US 174 | - Subsection: politics 175 | - Story: News 176 | - Best case prediction: The artile will talk about Bernie Sander, Democrats and Fox News in 2020 election. 177 | 178 | ### Result 179 | 180 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 181 | |---|---|---|---|---|---|---| 182 | | [Kamala Harris Drops Out of the 2020 Presidential Race](https://www.nytimes.com/2019/12/03/us/politics/kamala-harris-campaign-drops-out.html) | 2 | Y | Y | Y | Kamala Harris | Unrelated | 183 | | [Elizabeth Warren Gains Ground in 2020 Field, One Plan at a Time](https://www.nytimes.com/2019/05/28/us/politics/elizabeth-warren-2020.html) | 1 | Y | Y | Y | Elizabeth Warren | Unrelated | 184 | | [Democrats, Don’t Abandon Fox News](https://www.nytimes.com/2019/03/08/opinion/fox-news-democrats-debate.html) | 7 | N | N | N | Democrats and Fox News | Very Related | 185 | 186 | ### Breakdown 187 | 188 | 1. Kamala Harris Drops Out of the 2020 Presidential Race 189 | 190 | - Tags: Kamala D Harris,2020 Election,Democrats,Campaign finance,Black People;African-Americans;Black Culture and History,Primaries,US Politics 191 | - Section: us 192 | - Subsection: politics 193 | - Story: News 194 | - Note: Mention about Sanders and Democrats 195 | 196 | 2. Elizabeth Warren Gains Ground in 2020 Field, One Plan at a Time 197 | 198 | - Tags: Elizabeth Warren,Campaign finance,2020 Election,US Politics,Primaries,Polls,Democrats,Iowa 199 | - Section: us 200 | - Subsection: politics 201 | - Story: News 202 | - Note: Mention about Democrats 203 | 204 | 3. Opinion | Democrats, Don’t Abandon Fox News 205 | 206 | - Tags: Democrats,Fox News Channel,Presidential debates;Democratic debates;2020 debates,2020 Election,News media;journalism,US Politics 207 | - Section: opinion 208 | - Subsection: none 209 | - Story: Informational 210 | -------------------------------------------------------------------------------- /tf-idf/README.md: -------------------------------------------------------------------------------- 1 | # Experiment 2 | 3 | 5 articles were compared against 33,914 articles with TF-IDF similarity. 4 | 5 | # [How My Worst Date Ever Became My Best](https://www.nytimes.com/2020/02/14/style/modern-love-worst-date-of-my-life-became-best.html) 6 | 7 | - Theme: Dating, divorced 8 | - Tags: Dating 9 | - Section: style 10 | - Subsection: modern love 11 | - Story: Human interest 12 | - best case prediction: The article will be about dating for older couples who have experienced marriage. 13 | 14 | ### Result 15 | 16 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 17 | |---|---|---|---|---|---|---| 18 | | [Telling the Truth Wasn’t an Option](https://www.nytimes.com/2020/03/20/style/modern-love-undocumented-immigrant-telling-truth-wasnt-option.html) | 1 | Y | Y | Y | Illegal Immigration | Related | 19 | | [Live Alone? We Want to Hear From You](https://www.nytimes.com/2020/04/24/style/modern-love-coronavirus-alone-callout.html) | 0 | Y | Y | N | Viewer request | Unrelated | 20 | | [Love in the Time of Low Expectations](https://www.nytimes.com/2020/01/24/style/modern-love-in-the-time-of-low-expectations.html) | 1 | Y | Y | Y | Dating | Related | 21 | 22 | 1. Telling the Truth Wasn’t an Option 23 | 24 | - Tags: Love,Dating,Illegal Immigration;Undocumented Migrants 25 | - Section: style 26 | - Subsection: modern love 27 | - Story: Human interest 28 | 29 | 2. Live Alone? We Want to Hear From You 30 | 31 | - Tags: Love,Coronavirus;COVID-19 32 | - Section: style 33 | - Subsection: modern love 34 | - Story: Informative (submission request) 35 | 36 | 3. Love in the Time of Low Expectations 37 | 38 | - Tags: Love,Dating 39 | - Section: style 40 | - Subsection: modern love 41 | - Story: Human interest 42 | 43 | # [A Deep-Sea Magma Monster Gets a Body Scan](https://www.nytimes.com/2019/12/03/science/axial-volcano-mapping.html) 44 | 45 | - Theme: Nature, 3D scan 46 | - Tags: Axial Seamount,Volcano,Oceans and Seas,Geology,3-D,Ships and Shipping,Sensor,Scanners,Pacific Ocean,Oregon,your-feed-science 47 | - Section: science 48 | - Subsection: none 49 | - Story: Informational 50 | - best case prediction: The article will talk about ocean and 3d scanning. 51 | 52 | ### Result 53 | 54 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 55 | |---|---|---|---|---|---|---| 56 | | [A 3D Encounter With a Violent Volcano’s Underbelly](https://www.nytimes.com/2019/12/18/science/volcano-3d-reunion-island.html) | 4 | Y | N | Y | 3D Mapped Volcano | Highly Related | 57 | | [Pressure, and Mystery, on the Rise](https://www.nytimes.com/2015/01/06/science/predicting-what-a-volcano-may-or-may-not-do-is-as-tricky-as-it-is-crucial-as-iceland-well-knows.html) | 1 | Y | Y | Y | Iceland's Volcano | Related | 58 | | [It’s Not Just Hawaii: The U.S. Has 169 Volcanoes That Could Erupt](https://www.nytimes.com/2018/05/14/us/us-active-volcanoes-hawaii.html) | 2 | N | N | Y | Volcanos | Related | 59 | 60 | ### Breakdown 61 | 62 | 1. A 3D Encounter With a Violent Volcano’s Underbelly 63 | 64 | - Tags: Volcano,Reunion Island,3-D,Helicopter,Research,Geology,Scientific Reports Journal,Marc Dumont,your-feed-science 65 | - Section: science 66 | - Subsection: TRILOBITES 67 | - Story: Informational 68 | 69 | 2. Pressure, and Mystery, on the Rise 70 | 71 | - Tags: Volcano,Iceland 72 | - Section: science 73 | - Subsection: none 74 | - Story: Informational 75 | 76 | 3. It’s Not Just Hawaii: The U.S. Has 169 Volcanoes That Could Erupt 77 | 78 | - Tags: Volcano,Geology,Emergency Response and Preparedness,Alaska Volcano Observatory,US Geological Survey,Aleutian Islands,Lassen Volcanic National Park, Calif.,Kilauea volcano,Mountain States,Alaska,Hawaii,Wyoming,Western States,Yellowstone National Park,Oregon,Washington 79 | - Section: us 80 | - Subsection: none 81 | - Story: Informational 82 | 83 | # [Renault and Nissan Try a New Way After Years When Carlos Ghosn Ruled](https://www.nytimes.com/2019/11/29/business/renault-nissan-mitsubishi-alliance.html) 84 | 85 | - Theme: Renault, Nissan and Ghosn 86 | - Tags: Cars,Appointments and Executive Changes,Nissan Motor,Renault,Carlos Ghosn,Jean-Dominique Senard,Makoto Uchida,Mitsubishi Motor 87 | - Section: buisness 88 | - Subsection: none 89 | - Story: News 90 | - Best case prediction: The artile will talk about what has happened to either Renault and Nissan in relation to Carlos Ghosn incident. 91 | 92 | ### Result 93 | 94 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 95 | |---|---|---|---|---|---|---| 96 | | [Nissan CEO Says 'No Merit' in Merger With Renault-Nikkei](https://www.nytimes.com/reuters/2018/04/25/business/25reuters-renault-nissan-m-a.html) | 3 | N | N | Y | Nissan and Renault | Related | 97 | | [Carlos Ghosn and the Roots of Nissan’s Decline](https://www.nytimes.com/2020/01/15/automobiles/nissan-carlos-ghosn-strategy.html) | 3 | N | N | N | Carlos Ghosn | Related | 98 | | [Renault Chooses Volkswagen Executive as New C.E.O.](https://www.nytimes.com/2020/01/28/business/renault-ceo-luca-de-meo.html) | 5 | Y | Y | Y | Renault CEO | Very Related | 99 | 100 | ### Breakdown 101 | 102 | 1. Nissan CEO Says 'No Merit' in Merger With Renault-Nikkei 103 | 104 | - Tags: US;RENAULT;NISSAN;M;A;Mergers / Acquisitions / Takeovers;Company News;Auto and Truck Manufacturers (TRBC);Japan;Western Europe;Equity Investments;France;Asia / Pacific;Restructuring / Reorganization;Europe;Company Management Issues 105 | - Section: japan 106 | - Subsection: none 107 | - Story: News 108 | - Note: the story had been taken down in NYT. It's originally published from Reuters. 109 | 110 | 2. Carlos Ghosn and the Roots of Nissan’s Decline 111 | 112 | - Tags: Cars,Sales and Discounts,Nissan Motor,Carlos Ghosn,US,Incentive program 113 | - Section: automobiles 114 | - Subsection: none 115 | - Story: Informational 116 | 117 | 3. Renault Chooses Volkswagen Executive as New C.E.O. 118 | 119 | - Tags: Cars,Appointments and Executive Changes,Nissan Motor,Renault,Volkswagen,Jean-Dominique Senard,France,Luca de Meo 120 | - Section: business 121 | - Subsection: none 122 | - Story: News 123 | 124 | # [Dominic Thiem Beats Rafael Nadal in Australian Open Quarterfinal](https://www.nytimes.com/2020/01/29/sports/tennis/thiem-nadal-australian-open.html) 125 | 126 | - Theme: Tennis, Dominic Thiem, Australian Open 127 | - Tags: Tennis,Australian Open,Novak Djokovic,Dominic Thiem 128 | - Section: sports 129 | - Subsection: tennis 130 | - Story: News 131 | - Best case prediction: The artile will talk about Dominic Thiem and Australian Open. 132 | 133 | ### Result 134 | 135 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 136 | |---|---|---|---|---|---|---| 137 | | [Novak Djokovic Wins the Australian Open](https://www.nytimes.com/2020/02/02/sports/tennis/australian-open-djokovic-thiem.html) | 4 | Y | Y | Y | Dominic vs Novak in Australian Open | Very Related | 138 | | [Thiem Warms Up for French Open With Victory in Lyon](https://www.nytimes.com/aponline/2018/05/26/sports/tennis/ap-ten-lyon-open.html) | 1 | N | N | Y | French Open about Thiem | Related | 139 | | [Deja Vu for French Open as Nadal Favorite Again](https://www.nytimes.com/reuters/2018/05/24/sports/tennis/24reuters-tennis-frenchopen-men.html) | 1 | Y | N | Y | French Open | Unrelated | 140 | 141 | ### Breakdown 142 | 143 | 1. Novak Djokovic Wins the Australian Open 144 | 145 | - Tags: Novak Djokovic,Dominic Thiem,Australian Open,Tennis 146 | - Section: sports 147 | - Subsection: tennis 148 | - Story: News 149 | 150 | 2. Thiem Warms Up for French Open With Victory in Lyon 151 | 152 | - Tags: Alexander Zverev,Lyon,Sports - Europe,France,Dominic Thiem,Rafael Nadal,Gilles Simon 153 | - Section: none 154 | - Subsection: none 155 | - Story: News 156 | - Note: From May, 2018. AP. 157 | 158 | 3. Deja Vu for French Open as Nadal Favorite Again 159 | 160 | - Tags: US,TENNIS,FRENCHOPEN,MEN,Germany,Previews,France,Spain,Sweden,Austria,Sport,Tennis,Europe,Nordic States,United Kingdom 161 | - Section: sports 162 | - Subsection: none 163 | - Story: News 164 | - Note: From May, 2018. Reuters. Mention Thiem. 165 | 166 | # [2020 Democrats Seek Voters in an Unusual Spot: Fox News](https://www.nytimes.com/2019/04/17/us/politics/fox-news-democrats-2020.html) 167 | 168 | - Theme: Bernie Sander, Democrat, Fox 169 | - Tags: Fox News Channel,Democrats,2020 Election,News media;journalism,TV,Bernard Sanders;Bernie Sanders,Presidential debates;Democratic debates;2020 debates 170 | - Section: US 171 | - Subsection: politics 172 | - Story: News 173 | - Best case prediction: The artile will talk about Bernie Sander, Democrats and Fox News in 2020 election. 174 | 175 | ### Result 176 | 177 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 178 | |---|---|---|---|---|---|---| 179 | | [Democrats, Don’t Abandon Fox News](https://www.nytimes.com/2019/03/08/opinion/fox-news-democrats-debate.html) | 7 | N | N | N | Democrats and Fox News | Very Related | 180 | | [The Nation: Perspective on the Press; Fox News Moves From the Margins to the Mainstream](https://www.nytimes.com/2002/12/01/weekinreview/nation-perspective-press-fox-moves-margins-mainstream.html) | 2 | N | N | Y | Fox News | Related | 181 | | [Handling of Donald Trump Puts Fox News on the Spot](https://www.nytimes.com/2015/08/12/business/media/handling-of-trump-puts-fox-on-the-spot.html) | 2 | N | N | Y | Trump and Fox (old) | Unrelated | 182 | 183 | ### Breakdown 184 | 185 | 1. Democrats, Don’t Abandon Fox News 186 | 187 | - Tags: Democrats,Fox News Channel,Presidential debates;Democratic debates;2020 debates,2020 Election,News media;journalism,US Politics 188 | - Section: opinion 189 | - Subsection: none 190 | - Story: Informational 191 | 192 | 2. The Nation: Perspective on the Press; Fox News Moves From the Margins to the Mainstream 193 | 194 | - Tags: Fox Entertainment Group,,TV 195 | - Section: opinion 196 | - Subsection: week in review 197 | - Story: Infomrational 198 | 199 | 3. Handling of Donald Trump Puts Fox News on the Spot 200 | 201 | - Tags: ,Fox News Channel,Donald Trump,2016 Presidential Election,Megyn Kelly,News media;journalism,Roger E Ailes,Rupert Murdoch 202 | - Section: business 203 | - Subsection: media 204 | - Story: News 205 | - NOte: No mention about Democrats. Old from 2015. 206 | -------------------------------------------------------------------------------- /doc2vec/README.md: -------------------------------------------------------------------------------- 1 | # Experiment 2 | 3 | 5 articles were compared against 33,914 articles with Doc2vec similarity. 4 | 5 | # [How My Worst Date Ever Became My Best](https://www.nytimes.com/2020/02/14/style/modern-love-worst-date-of-my-life-became-best.html) 6 | 7 | - Theme: Dating, divorced 8 | - Tags: Dating 9 | - Section: style 10 | - Subsection: modern love 11 | - Story: Human interest 12 | - best case prediction: The article will be about dating for older couples who have experienced marriage. 13 | 14 | ### Result 15 | 16 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 17 | |---|---|---|---|---|---|---| 18 | | ["After Losing My Sight, Struggling to Be Seen"](https://www.nytimes.com/2020/01/17/style/modern-love-after-losing-my-sight-struggling-to-be-seen-blindness.html) | 1 | Y | Y | Y | Dating and Service Dog | Related | 19 | | [My Quarantine Boyfriend Lost Everything (but Found Me)](https://www.nytimes.com/2020/06/05/style/modern-love-coronavirus-quarantine-boyfriend-lost-everything.html) | 1 | Y | Y | Y | Dating | Related | 20 | | [Together](https://www.nytimes.com/2020/06/12/style/modern-love-coronavirus-living-together.html) | 1 | Y | Y | Y | Family | Unrelated | 21 | 22 | 1. "After Losing My Sight, Struggling to Be Seen" 23 | 24 | - Tas: Love,Blindness,Dating,Writer,Service Dogs;Support Animals 25 | - Section: style 26 | - Subsection: modern love 27 | - Story: Human interest 28 | 29 | 2. My Quarantine Boyfriend Lost Everything (but Found Me) 30 | 31 | - Tas: Love,Dating,Coronavirus;COVID-19,NYC 32 | - Section: style 33 | - Subsection: modern love 34 | - Story: Human interest 35 | 36 | 3. Together 37 | 38 | - Tags: Quarantines;Social Distancing;Shelter at home,Coronavirus;COVID-19,Dating,Family,Love 39 | - Section: style 40 | - Subsection: modern love 41 | - Story: Human interest 42 | 43 | # [A Deep-Sea Magma Monster Gets a Body Scan](https://www.nytimes.com/2019/12/03/science/axial-volcano-mapping.html) 44 | 45 | - Theme: Nature, 3D scan 46 | - Tags: Axial Seamount,Volcano,Oceans and Seas,Geology,3-D,Ships and Shipping,Sensor,Scanners,Pacific Ocean,Oregon,your-feed-science 47 | - Section: science 48 | - Subsection: none 49 | - Story: Informational 50 | - best case prediction: The article will talk about ocean and 3d scanning. 51 | 52 | ### Result 53 | 54 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 55 | |---|---|---|---|---|---|---| 56 | | [A 3D Encounter With a Violent Volcano’s Underbelly](https://www.nytimes.com/2019/12/18/science/volcano-3d-reunion-island.html) | 4 | Y | N | Y | 3D Mapped Volcano | Highly Related | 57 | | [The Mount St. Helens Eruption Was the Volcanic Warning We Needed](https://www.nytimes.com/2020/05/18/science/mt-st-helens-eruption.html) | 3 | Y | Y | N | Volcano | Related | 58 | | [The Loyal Engineers Steering NASA’s Voyager Probes Across](https://www.nytimes.com/2017/08/03/magazine/the-loyal-engineers-steering-nasas-voyager-probes-across-the-universe.html) | 0 | Y | Y | N | NASA | Unrelated | 59 | 60 | ### Breakdown 61 | 62 | 1. A 3D Encounter With a Violent Volcano’s Underbelly 63 | 64 | - Tags: Volcano,Reunion Island,3-D,Helicopter,Research,Geology,Scientific Reports Journal,Marc Dumont,your-feed-science 65 | - Section: science 66 | - Subsection: TRILOBITES 67 | - Story: Informational 68 | 69 | 2. The Mount St. Helens Eruption Was the Volcanic Warning We Needed 70 | 71 | - Tags: your-feed-science,Volcano,Geology,Mount St Helens,Washington 72 | - Section: science 73 | - Subsection: none 74 | - Story: Past events 75 | 76 | 3. The Loyal Engineers Steering NASA’s Voyager Probes Across 77 | 78 | - Tags: NASA,Space,Voyager program 79 | - Section: science 80 | - Subsection: none 81 | - Story: Human interest 82 | 83 | # [Renault and Nissan Try a New Way After Years When Carlos Ghosn Ruled](https://www.nytimes.com/2019/11/29/business/renault-nissan-mitsubishi-alliance.html) 84 | 85 | - Theme: Renault, Nissan and Ghosn 86 | - Tags: Cars,Appointments and Executive Changes,Nissan Motor,Renault,Carlos Ghosn,Jean-Dominique Senard,Makoto Uchida,Mitsubishi Motor 87 | - Section: buisness 88 | - Subsection: none 89 | - Story: News 90 | - Best case prediction: The artile will talk about what has happened to either Renault and Nissan in relation to Carlos Ghosn incident. 91 | 92 | ### Result 93 | 94 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 95 | |---|---|---|---|---|---|---| 96 | | [Renault Chooses Volkswagen Executive as New C.E.O.](https://www.nytimes.com/2020/01/28/business/renault-ceo-luca-de-meo.html) | 5 | Y | Y | Y | Renault CEO | Very Related | 97 | | [Renault and Nissan Unveil Plan to Patch Tattered Alliance](https://www.nytimes.com/2020/05/27/business/renault-nissan-alliance.html) | 4 | Y | Y | Y | Nissan and Renault | Very Related | 98 | | [Fiat Chrysler and Peugeot Agree on Terms to Forge New Auto Giant](https://www.nytimes.com/2019/12/18/business/fiat-chrysler-peugeot.html) | 2 | Y | Y | Y | Car Industry | Unrelated | 99 | 100 | ### Breakdown 101 | 102 | 1. Renault Chooses Volkswagen Executive as New C.E.O. 103 | 104 | - Tags: Cars,Appointments and Executive Changes,Nissan Motor,Renault,Volkswagen,Jean-Dominique Senard,France,Luca de Meo 105 | - Section: business 106 | - Subsection: none 107 | - Story: News 108 | 109 | 2. Renault and Nissan Unveil Plan to Patch Tattered Alliance 110 | 111 | - Tags: Cars,Nissan Motor,Renault,Electric Cars and Hybrids,Carlos Ghosn,Jean-Dominique Senard,Coronavirus;COVID-19 112 | - Section: business 113 | - Subsection: none 114 | - Story: News 115 | 116 | 3. Fiat Chrysler and Peugeot Agree on Terms to Forge New Auto Giant 117 | 118 | - Tags: Cars,Mergers and Acquisitions,,GM,Peugeot,John Elkann,Carlos Tavares,France 119 | - Section: business 120 | - Subsection: none 121 | - Story: News 122 | - Note: Not related to Nissan. No mention of it. Only Renault and French government. 123 | 124 | # [Dominic Thiem Beats Rafael Nadal in Australian Open Quarterfinal](https://www.nytimes.com/2020/01/29/sports/tennis/thiem-nadal-australian-open.html) 125 | 126 | - Theme: Tennis, Dominic Thiem, Australian Open 127 | - Tags: Tennis,Australian Open,Novak Djokovic,Dominic Thiem 128 | - Section: sports 129 | - Subsection: tennis 130 | - Story: News 131 | - Best case prediction: The artile will talk about Dominic Thiem and Australian Open. 132 | 133 | ### Result 134 | 135 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 136 | |---|---|---|---|---|---|---| 137 | | [Djokovic vs. Federer, a Rivalry for the Ages, Is One-Sided This Time](https://www.nytimes.com/2020/01/30/sports/tennis/djokovic-federer-australian-open.html) | 3 | Y | Y | Y | Australian Open | Related | 138 | | [Roger Federer Rallies to Reach the Final in Indian Wells](https://www.nytimes.com/2018/03/17/sports/roger-federer-indian-wells.html) | 1 | Y | N | Y | Paribas Open and Rogeer Federer | Unrelated | 139 | | [Alexander Zverev and Johanna Konta Lose in French Open Upsets](https://www.nytimes.com/2017/05/30/sports/tennis/french-open-day-3-zverev.html) | 1 | Y | Y | Y | French Open | Unrelated | 140 | 141 | ### Breakdown 142 | 143 | 1. Djokovic vs. Federer, a Rivalry for the Ages, Is One-Sided This Time 144 | 145 | - Tags: Tennis,Australian Open,Novak Djokovic,Roger Federer 146 | - Section: sports 147 | - Subsection: tennis 148 | - Story: News 149 | - Note: Same on Australian Open. Not about Dominic 150 | 151 | 2. Roger Federer Rallies to Reach the Final in Indian Wells 152 | 153 | - Tags: Tennis,Indian Wells Masters;BNP Paribas Open,Roger Federer,Borna Coric 154 | - Section: sports 155 | - Subsection: none 156 | - Story: News 157 | - Note: from 2018. No mention about Dominic 158 | 159 | 3. Alexander Zverev and Johanna Konta Lose in French Open Upsets 160 | 161 | - Tags: French Open,Tennis,Johanna Konta,Alexander Zverev,Juan Martin del Potro,Fernando Verdasco 162 | - Section: sports 163 | - Subsection: tennis 164 | - Story: News 165 | - Note: from 2017. No mention about Dominic 166 | 167 | # [2020 Democrats Seek Voters in an Unusual Spot: Fox News](https://www.nytimes.com/2019/04/17/us/politics/fox-news-democrats-2020.html) 168 | 169 | - Theme: Bernie Sander, Democrat, Fox 170 | - Tags: Fox News Channel,Democrats,2020 Election,News media;journalism,TV,Bernard Sanders;Bernie Sanders,Presidential debates;Democratic debates;2020 debates 171 | - Section: US 172 | - Subsection: politics 173 | - Story: News 174 | - Best case prediction: The artile will talk about Bernie Sander, Democrats and Fox News in 2020 election. 175 | 176 | ### Result 177 | 178 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 179 | |---|---|---|---|---|---|---| 180 | | [Elizabeth Warren’s Allies Claim ‘Erasure’ As They Seek To Reignite Campaign](https://www.nytimes.com/2020/02/18/us/politics/elizabeth-warren-nevada.html) | 5 | Y | Y | Y | Elizabeth's debate | Unrelated | 181 | | [Did America Misjudge Bernie Sanders? Or Did He Misjudge America?](https://www.nytimes.com/2020/03/16/magazine/bernie-sanders-campaign.html) | 4 | Y | Y | Y | Bernie Sanders | Related | 182 | | [Frozen in Anxiety: How Democratic Leaders Struggled to Confront Bernie Sanders](https://www.nytimes.com/2020/03/02/us/politics/bernie-sanders-democratic-party.html) | 5 | Y | Y | N | Bernie Sanders | Related | 183 | 184 | ### Breakdown 185 | 186 | 1. Elizabeth Warren’s Allies Claim ‘Erasure’ As They Seek To Reignite Campaign 187 | 188 | - Tags: 2020 Election,Elizabeth Warren,Democrats,US Politics,Primaries,Presidential debates;Democratic debates;2020 debates,Campaign finance,Voting,Polls 189 | - Section: us 190 | - Subsection: politics 191 | - Story: News 192 | - Note: No fox 193 | 194 | 2. Did America Misjudge Bernie Sanders? Or Did He Misjudge America? 195 | 196 | - Tags: Bernard Sanders;Bernie Sanders,2020 Election,Democrats,Joe Biden,Socialism,Vermont,US Politics 197 | - Section: us 198 | - Subsection: politics 199 | - Story: Personality 200 | - Note: No fox. 201 | 202 | 3. Frozen in Anxiety: How Democratic Leaders Struggled to Confront Bernie Sanders 203 | 204 | - Tags: 2020 Election,US Politics,Democrats,Primaries,Bernard Sanders;Bernie Sanders,Joe Biden,Mike Bloomberg,Amy Klobuchar,Elizabeth Warren,Donald Trump 205 | - Section: us 206 | - Subsection: politics 207 | - Story: Personality 208 | -------------------------------------------------------------------------------- /bert/README.md: -------------------------------------------------------------------------------- 1 | # Experiment 2 | 3 | 5 articles were compared against 33,914 articles with Bidirectional Encoder Representations from Transformers (BERT) similarity. 4 | 5 | # [How My Worst Date Ever Became My Best](https://www.nytimes.com/2020/02/14/style/modern-love-worst-date-of-my-life-became-best.html) 6 | 7 | - Theme: Dating, divorced 8 | - Tags: Dating 9 | - Section: style 10 | - Subsection: modern love 11 | - Story: Human interest 12 | - best case prediction: The article will be about dating for older couples who have experienced marriage. 13 | 14 | ### Result 15 | 16 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 17 | |---|---|---|---|---|---|---| 18 | | [Why Are All the Exes Texting?](https://www.nytimes.com/2020/05/29/style/modern-love-coronavirus-why-are-all-the-exes-texting.html) | 1 | Y | Y | Y | Dating | Related | 19 | | [When Love Seems Too Easy to Trust](https://www.nytimes.com/2018/04/13/style/when-love-seems-too-easy-to-trust.html) | 1 | Y | Y | Y | Dating | Related | 20 | | [He Saved His Last Lesson for Me](https://www.nytimes.com/2020/03/06/style/modern-love-long-distance-dating-korea.html) | 1 | Y | Y | Y | Dating | Related | 21 | 22 | 1. Why Are All the Exes Texting? 23 | 24 | - Tags: Dating,Quarantines;Social Distancing;Shelter at home,Coronavirus;COVID-19 25 | - Section: style 26 | - Subsection: modern love 27 | - Story: human interest 28 | 29 | 2. When Love Seems Too Easy to Trust 30 | 31 | - Tags: Love,Dating,Movies,NYC,When Harry Met Sally 32 | - Section: style 33 | - Subsection: modern love 34 | - Story: Human interest 35 | 36 | 3. He Saved His Last Lesson for Me 37 | 38 | - Tags: Love,Dating 39 | - Section: style 40 | - Subsection: modern love 41 | - Story: human interest 42 | 43 | # [A Deep-Sea Magma Monster Gets a Body Scan](https://www.nytimes.com/2019/12/03/science/axial-volcano-mapping.html) 44 | 45 | - Theme: Nature, 3D scan 46 | - Tags: Axial Seamount,Volcano,Oceans and Seas,Geology,3-D,Ships and Shipping,Sensor,Scanners,Pacific Ocean,Oregon,your-feed-science 47 | - Section: science 48 | - Subsection: none 49 | - Story: Informational 50 | - best case prediction: The article will talk about ocean and 3d scanning. 51 | 52 | ### Result 53 | 54 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 55 | |---|---|---|---|---|---|---| 56 | | [As Lava Oozes, Hawaii Officials Say Rest of Island Is Open](https://www.nytimes.com/aponline/2018/05/11/us/ap-us-hawaii-volcano.html) | 0 | N | N | N | Hawaii Volcano Warning | Related | 57 | | [Did Heavy Rain Make Hawaii’s Kilauea Volcano Erupt?](https://www.nytimes.com/2020/04/22/science/kilauea-volcano-eruption-rain.html) | 3 | Y | Y | N | Hawaii Volcano | Related | 58 | | [A 3D Encounter With a Violent Volcano’s Underbelly](https://www.nytimes.com/2019/12/18/science/volcano-3d-reunion-island.html) | 4 | Y | N | Y | 3D Mapped Volcano | Highly Related | 59 | 60 | ### Breakdown 61 | 62 | 1. As Lava Oozes, Hawaii Officials Say Rest of Island Is Open 63 | 64 | - Tags: AP Top News,Hawaii,Kilauea,North America,Business,Science,Travel,U.S. News 65 | - Section: none (AP) 66 | - Subsection: none (AP) 67 | - Story: News 68 | 69 | 2. Did Heavy Rain Make Hawaii’s Kilauea Volcano Erupt? 70 | 71 | - Tags: Kilauea volcano,Volcano,Hawaii,Rain,Nature Journal,Geology,Research,your-feed-science,Jamie Farquharson,Falk Amelung 72 | - Section: science 73 | - Subsection: none 74 | - Story: Past events 75 | 76 | 3. A 3D Encounter With a Violent Volcano’s Underbelly 77 | 78 | - Tags: Volcano,Reunion Island,3-D,Helicopter,Research,Geology,Scientific Reports Journal,Marc Dumont,your-feed-science 79 | - Section: science 80 | - Subsection: TRILOBITES 81 | - Story: Informational 82 | 83 | 84 | # [Renault and Nissan Try a New Way After Years When Carlos Ghosn Ruled](https://www.nytimes.com/2019/11/29/business/renault-nissan-mitsubishi-alliance.html) 85 | 86 | - Theme: Renault, Nissan and Ghosn 87 | - Tags: Cars,Appointments and Executive Changes,Nissan Motor,Renault,Carlos Ghosn,Jean-Dominique Senard,Makoto Uchida,Mitsubishi Motor 88 | - Section: buisness 89 | - Subsection: none 90 | - Story: News 91 | - Best case prediction: The artile will talk about what has happened to either Renault and Nissan in relation to Carlos Ghosn incident. 92 | 93 | ### Result 94 | 95 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 96 | |---|---|---|---|---|---|---| 97 | | [Renault and Nissan Unveil Plan to Patch Tattered Alliance](https://www.nytimes.com/2020/05/27/business/renault-nissan-alliance.html) | 4 | Y | Y | Y | Nissan and Renault | Very Related | 98 | | [Volkswagen's New CEO Treads Carefully on Road to Reform](https://www.nytimes.com/reuters/2018/04/13/business/13reuters-volkswagen-management.html) | 1 | N | N | Y | Volkswagen | Unrelated | 99 | | [Fiat Chrysler and Peugeot Agree on Terms to Forge New Auto Giant](https://www.nytimes.com/2019/12/18/business/fiat-chrysler-peugeot.html) | 2 | Y | Y | Y | Car Industry | Unrelated | 100 | 101 | ### Breakdown 102 | 103 | 1. Renault and Nissan Unveil Plan to Patch Tattered Alliance 104 | 105 | - Tags: Cars,Nissan Motor,Renault,Electric Cars and Hybrids,Carlos Ghosn,Jean-Dominique Senard,Coronavirus;COVID-19 106 | - Section: business 107 | - Subsection: none 108 | - Story: News 109 | 110 | 2. Volkswagen's New CEO Treads Carefully on Road to Reform 111 | 112 | - Tags: US;VOLKSWAGEN;MANAGEMENT;Germany;Labour Disputes;Key Personnel Changes;Company News;Auto and Truck Manufacturers (TRBC);Western Europe;Initial Public Offerings;New Issues;Italy;Directors Dealings;Major News;Restructuring / Reorganization;Europe;Company Management Issues;Labour / Personnel 113 | - Section: finance (Reuters) 114 | - Subsection: none (Reuters) 115 | - Story: News 116 | - Note: from 2018. 117 | 118 | 3. Fiat Chrysler and Peugeot Agree on Terms to Forge New Auto Giant 119 | 120 | - Tags: Cars,Mergers and Acquisitions,,GM,Peugeot,John Elkann,Carlos Tavares,France 121 | - Section: business 122 | - Subsection: none 123 | - Story: News 124 | - Note: Not related to Nissan. No mention of it. Only Renault and French government. 125 | 126 | # [Dominic Thiem Beats Rafael Nadal in Australian Open Quarterfinal](https://www.nytimes.com/2020/01/29/sports/tennis/thiem-nadal-australian-open.html) 127 | 128 | - Theme: Tennis, Dominic Thiem, Australian Open 129 | - Tags: Tennis,Australian Open,Novak Djokovic,Dominic Thiem 130 | - Section: sports 131 | - Subsection: tennis 132 | - Story: News 133 | - Best case prediction: The artile will talk about Dominic Thiem and Australian Open. 134 | 135 | ### Result 136 | 137 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 138 | |---|---|---|---|---|---|---| 139 | | [Novak Djokovic Wins the Australian Open](https://www.nytimes.com/2020/02/02/sports/tennis/australian-open-djokovic-thiem.html) | 4 | Y | Y | Y | Dominic vs Novak in Australian Open | Very Related | 140 | | [Djokovic vs. Federer, a Rivalry for the Ages, Is One-Sided This Time](https://www.nytimes.com/2020/01/30/sports/tennis/djokovic-federer-australian-open.html) | 3 | Y | Y | Y | Australian Open | Related | 141 | | [Though Federer Survives 2nd 5-Set Match, His Play Invites Questions](https://www.nytimes.com/2017/08/31/sports/tennis/roger-federer-us-open.html) | 1 | Y | Y | Y | US Open | Unrelated | 142 | 143 | ### Breakdown 144 | 145 | 1. Novak Djokovic Wins the Australian Open 146 | 147 | - Tags: Novak Djokovic,Dominic Thiem,Australian Open,Tennis 148 | - Section: sports 149 | - Subsection: tennis 150 | - Story: News 151 | 152 | 2. Djokovic vs. Federer, a Rivalry for the Ages, Is One-Sided This Time 153 | 154 | - Tags: Tennis,Australian Open,Novak Djokovic,Roger Federer 155 | - Section: sports 156 | - Subsection: tennis 157 | - Story: News 158 | - Note: Same on Australian Open. Not about Dominic 159 | 160 | 3. Though Federer Survives 2nd 5-Set Match, His Play Invites Questions 161 | 162 | - Tags: Tennis,US Open Tennis,Roger Federer,Mikhail Youzhny 163 | - Section: sports 164 | - Subsection: tennis 165 | - Story: News 166 | - Note: from 2017. US Open. No mention about Dominic. 167 | 168 | # [2020 Democrats Seek Voters in an Unusual Spot: Fox News](https://www.nytimes.com/2019/04/17/us/politics/fox-news-democrats-2020.html) 169 | 170 | - Theme: Bernie Sander, Democrat, Fox 171 | - Tags: Fox News Channel,Democrats,2020 Election,News media;journalism,TV,Bernard Sanders;Bernie Sanders,Presidential debates;Democratic debates;2020 debates 172 | - Section: US 173 | - Subsection: politics 174 | - Story: News 175 | - Best case prediction: The artile will talk about Bernie Sander, Democrats and Fox News in 2020 election. 176 | 177 | ### Result 178 | 179 | |Title|Tag Overlap|Section Overlap|Subsection Overlap|Style Overlap|Theme|Subjective| 180 | |---|---|---|---|---|---|---| 181 | | [Frozen in Anxiety: How Democratic Leaders Struggled to Confront Bernie Sanders](https://www.nytimes.com/2020/03/02/us/politics/bernie-sanders-democratic-party.html) | 5 | Y | Y | N | Bernie Sanders | Related | 182 | | [No Disclosure? No Problem. Sean Hannity Gets a Pass at Fox News.](https://www.nytimes.com/2018/04/17/business/media/sean-hannity-fox-news.html) | 2 | N | N | Y | Fox | Related | 183 | | [Bernie Sanders Had a Problem With MSNBC. Then Came Super Tuesday.](https://www.nytimes.com/2020/03/05/business/media/msnbc-bernie-sanders-media.html) | 7 | N | N | Y | Sanders and MSNBC | Very Related | 184 | 185 | ### Breakdown 186 | 187 | 1. Frozen in Anxiety: How Democratic Leaders Struggled to Confront Bernie Sanders 188 | 189 | - Tags: 2020 Election,US Politics,Democrats,Primaries,Bernard Sanders;Bernie Sanders,Joe Biden,Mike Bloomberg,Amy Klobuchar,Elizabeth Warren,Donald Trump 190 | - Section: us 191 | - Subsection: politics 192 | - Story: Personality 193 | 194 | 2. No Disclosure? No Problem. Sean Hannity Gets a Pass at Fox News. 195 | 196 | - Tags: Sean Hannity,Michael Cohen,Fox News Channel,TV,Ratings,Donald Trump 197 | - Section: business 198 | - Subsection: media 199 | - Story: News 200 | - Note: from 2018. No Bernie. 201 | 202 | 3. Bernie Sanders Had a Problem With MSNBC. Then Came Super Tuesday. 203 | 204 | - Tags: Bernard Sanders;Bernie Sanders,Democrats,2020 Election,MSNBC,News media;journalism,Primaries,TV,Rachel Maddow,Chris Matthews 205 | - Section: business 206 | - Subsection: media 207 | - Story: News 208 | - Note: Very good. Another right media appearance by Democrat 209 | --------------------------------------------------------------------------------