Schedule

Make sure to reload this page to ensure you’re seeing the latest version. Readings should be done before the indicated class. Texts include:

• JM = Jurafsky and Martin, Speech and Language Processing, 3rd edition draft chapters
• INLP Eisenstein, Intro to NLP, online text (or book version)

Lecture videos are available on the Piazza Resources page.

Week 1 (8/25, 8/27): Introduction

• HW0 is out. Due on Gradescope next Friday (9/4).
• "Quiz 0" survey is out. Please complete over the weekend. It's quick and easy!

Week 2 (9/1, 9/3): Words, Regexes, and Unix coding session

• Reading: JM 2.1-2.4
• Reading: Ken Church, "Unix for Poets"
• For Thursday, make sure to have unix terminal/shell available on your computer, and to have read Church and JM 2.4.
• Unix processing exercise - from 9/3 lecture

Week 3 (9/8, 9/10): Text Classification (Naive Bayes, sentiment analysis, evaluation)

• Reading: JM ch. 4, "Naive Bayes and Sentiment Classification"
• Optional: Pang et al., 2002, "Thumbs up? Sentiment Classification using Machine Learning Techniques".
  The original sentiment classification paper is still a great introduction! One update: what they call "maxent" we now call "logistic regression," and the differences between logreg versus linear SVMs are now typically minor.
• 9/10 keyword sentiment demo: ipynb, html

Week 4 (9/15, 9/17): Projects / More Classifiers

• Project page
• Reading: JM ch. 5, Logistic Regression (only 5.1, 5.5-5.7)

Week 5 (9/22, 9/24): Annotations and Evaluation

• Reading: INLP 4.5, "Building datasets" for annotations
• Reading: INLP 4.4, "Evaluating classifiers" for evaluation
• Also discuss stat testing on 9/29

Week 6 (9/29, 10/1): Rules and Classification in Context

• 9/29 slides on VADER. Paper: Hutto and Gilbert 2014, "VADER: A Parsimonious Rule-based Model for Sentiment Analysis of Social Media Text".

Week 7 (10/6, 10/8): Word Embeddings

• Reading: JM ch. 6, Vector Semantics/Embeddings
• Word embedding demo [html] [ipynb]

Week 8 (10/13, 10/15): Midterm review and midterm (10/15)

• Practice midterm questions and partial solutions on Piazza Resources page.
• Midterm released 9am on 10/15; finish anytime between then and 9pm.

Week 9 (10/20, 10/22): Neural networks and BERT

• Reading: INLP ch. 3, Nonlinear classification
• See also: JM ch. 7, Neural Networks
• See also a zillion online resources on neural language models and BERT. I am not sure what's a good review for the breadth of BERT applications, but to review what the model does I'd suggest:
  • Rogers et al. 2020, A Primer in BERTology: What we know about how BERT works
  • If you want more on the Transformer architecture, see this, this, and the Collins video linked in the slides.

Week 10

• 10/27: Project questions
• 10/19: Clinical NLP (Guest lecture: Rumeng Li)

Week 11

• 11/3: Ethics in NLP (Guest lecture: Su Lin Blodgett)
• 11/5: Machine translation

Week 12

• 11/10: Generation (Guest lecture: Nader Akoury)
• 11/12: Social Data Science with Text (Guest lecture: Katie Keith)

Week 13

• 11/17: Project presentations, day 1
• 11/19: Project presentations, day 2

Final project reports are due December 4 (last day of final exam period). Late days cannot be used.