'Discovering Laws of Semantic Change' and 'Extracting Social and Pragmatic Meaning from Everyday Interactions:' On Applying Computational Linguistics to the Social Science of Language

Computational linguistics has an important role to play in helping uncover the historical and social context of the language we use. I first describe our work on understanding how word meaning changes over time, using computational models based on vector semantics to quantify semantic change in historical corpora. We propose two quantitative laws of semantic change and test them over a 200 year window in four languages: (1) the rate of semantic change scales with an inverse power-law of word frequency, and (2) words that are more polysemous have higher rates of semantic change.  I then turn to our research on extracting meaning from everyday language.  I show how the rhetorical structure that authors use in writing their scientific abstracts predicts the future popularity of scientific areas, and discuss the way economic, social, and psychological variables are reflected in the language we use to talk about food in menus and restaurant reviews. I conclude with our most recent work that applies computational models of respect (based on theories of positive and negative face) to the language in police body-camera footage. Our results suggest that police use different respect levels when talking to community members of different races, even in everyday interactions.

Dan Jurafsky is Professor and Chair of Linguistics and Professor of Computer Science at Stanford University. He is the recipient of a 2002 MacArthur Fellowship, is the co-author with Jim Martin of the widely-used textbook "Speech and Language Processing,” and co-created with Chris Manning one of the first massively open online courses, Stanford's course in Natural Language Processing. His new trade book "The Language of Food: A Linguist Reads the Menu" came out on September 15, 2014, and was a finalist for the 2015 James Beard Award. Dan received a B.A in Linguistics in 1983 and a Ph.D. in Computer Science in 1992 from the University of California at Berkeley, was a postdoc 1992-1995 at the International Computer Science Institute. His research ranges widely across computational linguistics; special interests include natural language understanding, machine translation, spoken language and conversation, the relationship between human and machine processing, and the application of natural language processing to the social and behavioral sciences.