| HW | Status | Task |
|---|---|---|
| HW1 | Complete | Implemented Regular Expression to filter all phone number and email address |
| HW2 | Complete | Implemented Laplace Unigram / Bigram, Stupid Backoff, Kneser-Ney Smoothing language model to complete the Autocorrect task, reached expected 25% accuracy |
| HW3 | Complete | Implemented Naive Bayes Algorithm to predict the sentiment of movie reviews, reached expected 80% accuracy |
| HW4 | Partially Complete | Implemented Boolean Naive Bayes and Sentiment Lexicon to predict movie review sentiments |
| HW5 | Complete | Extracted the features from names, implemented the Word shape generalization. Reached the expected f1 score: 0.85 |
| HW6 | Complete | Implemented the Propabilitic Context Free Grammar features. Expected Perplexity lower than 400, achieved 368 |
| HW7 | lose points | Implemented the CKY algorithm, and implemented the vertical markovization |
| HW8 | Complete | Implemented the Inverted Index, Boolean Retrieval system, TF-IDF scoring and Cosine Similarity algorithm. Pass all tests. |
anisiomqs/NLP-Stanford-Assignments
The assignments from Stanford Natural Language Processing Course
Java