Colloquium Details
Machine Learning for Computational Social Science
Speaker: Jacob Eisenstein, Georgia Institute of Technology
Location: 60 Fifth Avenue 150
Date: March 30, 2018, 11 a.m.
Host: Rob Fergus
Synopsis:
Our social, personal, and political lives are increasingly mediated by technology. This change has introduced new problems, such as echo chambers and viral hoaxes. But it has also brought exciting new opportunities to understand the social world, using data and methods that earlier social scientists could only dream of. The first generation of computational social science focused on sensing technologies and social network analysis; the next generation will be driven by artificial intelligence, which makes it possible to operationalize social science constructs such as influence, attention, formality, and respect. In this talk, I will present an approach to computational social science that leverages customized machine learning models of heterogeneous data, including language, social networks, and spatiotemporal cascades. First, I will show how unsupervised machine learning over social network labelings and text makes it possible to induce the social meanings of address terms such as "Ms" and "dude". Next, I will describe how the spread of linguistic innovations can serve as evidence for sociocultural affinity and influence, using Bayesian vector autoregressive models and the Hawkes process. Finally, I will present recent research analyzing the causal impact of closing forums for hate speech.
Speaker Bio:
Jacob Eisenstein is an Associate Professor in the School of Interactive Computing at Georgia Tech. He works on computational sociolinguistics, social media analysis, and machine learning. He is a recipient of the NSF CAREER Award, a member of the Air Force Office of Scientific Research (AFOSR) Young Investigator Program, and was a SICSA Distinguished Visiting Fellow at the University of Edinburgh. His work has also been supported by the National Institutes for Health, the National Endowment for the Humanities, and Google. Jacob was a Postdoctoral researcher at Carnegie Mellon and the University of Illinois. He completed his Ph.D. at MIT in 2008, winning the George M. Sprowls dissertation award. Jacob's research has been featured in the New York Times, National Public Radio, and the BBC.
Notes:
In-person attendance only available to those with active NYU ID cards.