Graduate school course in Natural Language Processing - weekly assignments.

Week 1 - Introduction To Text As Data

• What is a 'Document Frequency Matrix' and what is it used for?

Week 2 - Text Pre-Processing

• Exploring best practices when pre-processing text data: tf-idf, tokenization, stop-words, stemming, etc.

Week 3 - Getting Data and APIs

• Exploring the use of 4 APIs: NYT, Steam, Facebook and Twitter. Used NYT API to preform a sentiment analysis on 6-months-worth of headlines via pytorch.

Week 4 - Methods AND Models

• Exploring regex and various models.

Week 5 - Dictionary-Based Methods

• Sentiment Analysis, Hyper-Parameter Tuning, Training Dataset Selection

Week 6 - Topic Models

• Exploring topic models

Week 7 - Supervised Learning Models

Week 8 - Unsupervised Learning/Clustering

Week 9 - Supervised Classification Models

Week 10 - Unsupervised Learning Methods

Week 11 - Word Embeddings

Week 12 - Course Project Submission