Noah Smith

Research interests:

natural language processing (and applications like machine translation, information extraction, question answering, and text-driven forecasting);
statistical machine learning (structured prediction, latent variables, sparsity, inference and optimization algorithms, and sometimes being Bayesian);
computational linguistics (morphology, syntax, and semantics); and
computational social science (sociolinguistics, political science, and economics).

Research group: Noah's ARK

Stuff I do:

I wrote a book about NLP and machine learning, published in 2011.
My South by Southwest panel on text-driven forecasting, 2011.
I'm on the editorial board of Computational Linguistics and the Journal of Artificial Intelligence Research and frequently serve on program committees for NLP and ML conferences (recently: ACL 2009, NAACL 2010, EMNLP 2010, ACL 2012).
I am the 2012–2013 secretary-treasurer of SIGDAT.

Prospective students: general advice about grad school and note to people who might be interested in joining my group.

Courses I taught/am teaching, and other tutorial materials

A few words from another, more famous, less hirsute Professor Smith. This image is copyright Jorge Cham; I really hope it's okay with Jorge for me to display it with a hyperlink (which you should follow for the rest of the joke and lots more Ph.D. fun).

Algorithms for Natural Language Processing (CMU, with Alon Lavie and Bob Frederking, Fall 2011): NLP course for graduate students. This course's next expected offering is Fall 2012.

Structured Prediction for Language and Other Discrete Data (CMU, with William Cohen, Fall 2011): advanced NLP/machine learning course for graduate students. This course's next expected offering is Fall 2013.

Natural Language Processing (CMU, Spring 2008, Spring 2009, Spring 2010, Spring 2011): introductory course for CS undergraduates. This course is being taught by a different instructor in Spring 2012.

Advanced Natural Language Processing Seminar (CMU, Spring 2009, Spring 2010, Spring 2011): readings in NLP and computational linguistics. This course's next expected offering is Fall 2012.

Of interest: Hidden Markov Models: All the Glorious Gory Details (October, 2004)

A handout for a guest lecture at U. Maryland: Log-Linear Models (December, 2004)

Predicting English (JHU CLSP summer workshop lab, Summers 2002 and 2003, taught by others since); see also our paper about this exercise.

My students and me, in April 2009. Picture by Mattt Thompson.

Noah Smith

Courses I taught/am teaching, and other tutorial materials

Curriculum vitae highlights