schedule.md

Schedule

All assignments and dates are subject to change

Readings will be added to the later weeks of the course in response to student interest and new developments in the field.

Readings are due on the day indicated.

Problem sets will be distributed via CMS no later than the Friday indicated. They are are due as indicated on CMS (often, but not always, two weeks from distribution).

Occasional brief responses are due as assigned on Tuesdays before 4:00pm. For details, see the Discussions section of Canvas.

• JM = Jurafsky and Martin, Speech and Language Processing, 3rd ed. (online)

Week	Monday	Wednesday	Friday
1 (8/21)	Introduction	Tokenization. Sentiment-aware tokenization Optional: JM, "Regular Expressions, Text Normalization, and Edit Distance"	Problem set 0: Setup and shakedown
2 (8/28)	Dictionary methods and vector space models. Reagan et al., "Emotional Arcs of Stories ..." Syuzhet package Optional: JM, "Lexicons for Sentiment, Affect, and Connotation"	Ramsay, "Algorithmic Criticism" (Canvas) Healy, "Fuck Nuance" Rambsy, #TheJayZMixtape	PS 1: Tokens, vectors, and regression
3 (9/4)	No class. Labor Day.	Regression. Davis, "Physical Traits that Define Men and Women in Literature" Ladd, "Understanding and Using Common Similarity Measures for Text Analysis". Introduction to regression
4 (9/11)	Clustering. Grimmer and Stewart, "Text as Data"	Moretti, "Slaughterhouse of Literature" (Canvas)	PS 2: Clustering and classification
5 (9/18)	Classification.	Allison et al., "Quantitative Formalism" Mauch et al., "The Evolution of Popular Music"
6 (9/25)	Feature importance and hypothesis testing. Monroe et al., "Fightin' Words" Sarkar, "The Importance of Human Interpretable Machine Learning" Confidence Intervals Hypothesis Testing	Underwood, "Why Literary Time is Measured in Minutes" Yauney et al., "Computational Prediction of Elapsed Narrative Time" Piper, "Characterization" (Canvas)
7 (10/2)	Topic models Boyd-Graber, Hu, and Mimno, Applications of Topic Models, chapters 1, 4, and 6. Altman, "The curse(s) of dimensionality"	Underwood, "The Life Cycles of Genres" Barron et al., "Individuals, Institutions, and Innovation in the Debates of the French Revolution"	No section meetings. Fall break. PS 3: Features and comparisons.
8 (10/9)	No class. Fall break.	NLP and feature expansion. Evans and Wilkens, "Nation, Ethnicity, and the Geography of British Fiction" Spacy 101 Optional: JM, ch. 8, "Sequence Labeling"
9 (10/16)	Static word embeddings. Mikolov et al., "Efficient Estimation of Word Representations in Vector Space" Ruder, "On Word Embeddings"	Nelson, "Leveraging the Alignment between Machine Learning and Intersectionality" (Canvas)	PS 4: Entities and static embeddings
10 (10/23)	BERT and contextual embeddings. Vaswani et al., "Attention is all you need" BERT for Humanists tutorials Optional: JM, ch. 7 (neural language models) and ch. 10 (transformers)	Thai et al., "RELIC: Retrieving Evidence for Literary Claims" Manjavacas and Fonteyn, "Models for Historical Languages"
11 (10/30)	Large language models and generative AI. Brown et al., "LLMs are Few-Shot Learners"	Norvig, "Chomsky and the Two Cultures of Statistical Learning"
12 (11/6)	Catch-up: Using BERT for classification	No class. Prof. Wilkens out of town.	PS 5: Contextual embeddings and LLMs
13 (11/13)	Catch-up: The BERT architecture "The Illustrated BERT" Optional: HuggingFace NLP/Transformers course	Text generation with LLMs Thai et al., "Literary Machine Translation"
14 (11/20)	LLM applications	No class. Thanksgiving.	No section meetings. Thanksgiving.
15 (11/27)	Multilingual NLP "mT5: A massively multilingual pre-trained text-to-text transformer"	Working with user-generated content. Antoniak, "Narrative Paths and Negotiation of Power in Birth Stories" Dym and Fiesler, "Ethical and Privacy Considerations for Research Using Online Fandom Data"	Review and exam preparation
16 (12/4)	Summary discussion and conclusions.	-----	-----