COMP550_19Fall

This is a repository for COMP 550: Natural Language Processing provied by McGill University in fall 2019.

Projects

• Sentiment_Analysis
  This is the folder for the third question of the first assignment. I planned to implement a sentiment analysis model using naive bayes, logistic regression, and support vector machine to classify a binary distributed dataset (either positive or negative). The dataset is the same one used in Seeing starts: Exploiting class relationships for sentiment categorization with respect to rating scale by Bo Pang and Lillian Lee in 2005. You may access the data from here. Also, the pretrained word embeddings, GloVe, is used in the experiment.

• Decipherment_with_HMM
  This is the folder for the third question of the second assignment. I planned to implement a standard hidden Markov model to solve a sets of ciphers encrypted in three different ways. The dataset is retrived from professor, which may not be able to be uploaded here. In addition to improve the tagging accuracy, I used nltk built-in corpus Penn Treebank to count the transition probilities.

References

1. Pang, Bo, and Lillian Lee. "Seeing stars: Exploiting class relationships for sentiment categorization with respect to rating scales." Proceedings of the 43rd annual meeting on association for computational linguistics. Association for Computational Linguistics, 2005.
2. Pennington, Jeffrey, Richard Socher, and Christopher Manning. "Glove: Global vectors for word representation." Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP). 2014.
3. Marcus, Mitchell, et al. "The Penn Treebank: annotating predicate argument structure." Proceedings of the workshop on Human Language Technology. Association for Computational Linguistics, 1994.

License

This project is under the MIT license.