This repo contains my coursework for NLP at USC. CS544 covered an overview of NLP topics such as information extraction, word embeddings, QA, NLI, machine translation, etc. We also gained a mastery of statistical (e.g., HMM, Naive Bayes) and state-of-the-art neural methods (e.g., BERT series, GRU, etc.) used in language technologies.
Here, I provide a very brief overview of hw's completed. For more info, see the specific README's in the respective folders.
- Naive Bayes
- Tasked to write a Naive Bayes classifier (from scratch) for sentiment analysis.
-
HW2 - Hidden markov models
- Develop a Hidden Markov Model part-of-speech tagger (from scratch) that works for any language.
-
HW3 - Ciphertext classification
- Tasked with a simple scenario for privacy-preserving NLU: developing a text classification model based on ciphertext.
-
HW4 - Precondition inference
- Develop a natural language reasoner to decide whether a precondition will enable or disable a statement.