Social Science and Text Analysis: Methods from the Frontier


February 5, 2018
12:30pm to 2:00pm
190 Doe Library
Get Directions


Karthik Ram and the Berkeley Institute for Data Science (BIDS) are pleased to announce that they will host a lunch/talk with Marti Hearst, Ken Benoit and Nick Adams on Feb 5th from 12:30-2pm.

Ken Benoit will discuss recent works modeling party positions as a function of the latent characteristics of political speech, and inducing relationships between left/right party positions and hand-labeled policy categories. Nick Adams will discuss his use of SVO-amplified LDA and crowd annotation to generate dynamic interactive models of protest and police activities during the Occupy movement. Marti Hearst, of Berkeley’s School of Information, will moderate the discussion and an audience Q&A about these talks and the broader topic of operationalizing social scientific concepts using text analysis techniques.

Bring an appetite for lunch... and for learning about methods from the text analysis frontier!

When: Monday, Feb 5th 12:30-2:00pm
Where: BIDS – 190 Doe Library
Register: Please RSVP so that lunch can be planned accurately.




Ken Benoit

Professor of Quantitative Social Research Methods, and Head of the Department of Methodology
London School of Economics and Political Science

Kenneth Benoit, Ph.D., is currently Professor of Quantitative Social Research Methods, and Head of the Department of Methodology at the London School of Economics and Political Science. His current research focuses on automated, quantitative methods for processing large amounts of textual data, mainly political texts and social media. Current interest span from the analysis of big data, including social media, and methods of text mining. His substantive research in political science focuses on comparative party competition, the European Parliament, electoral systems, and the effects of campaign spending. His other methodological interests include general statistical methods for the social sciences, especially those relating to measurement. Recent data large-scale measurement projects in which he has been involved include estimating policy positions of political parties through crowd-sourced data, expert surveys, manifesto coding, and text analysis.

Marti Hearst

Professor, School of Information and the EECS Department
University of California, Berkeley

Marti Hearst, Ph.D., is a professor in the School of Information and the EECS Department at UC Berkeley. Her primary research interests are user interfaces for search engines, information visualization, natural language processing, and improving MOOCs. She wrote the first book on Search User Interfaces. Prof. Hearst was named a Fellow of the ACM in 2013, a member of the CHI Academy in 2017, and has received an NSF CAREER award, an IBM Faculty Award, two Google Research Awards, an Okawa Foundation Fellowship, four Excellence in Teaching Awards, and has been principal investigator for more than $3.5M in research grants.

Nick Adams

Nick Adams, Ph.D., is the Founder and Director of the GoodlyLabs and a staff scientist at the UC Berkeley Institute for Data Science. His current research builds collaborative software that online communities can use to closely analyze textual data at large scales. In one project, teams of students and volunteers use GoodlyLabs’ TextThresher software to analyze protester and police interactions as revealed through 8,000 news accounts of nearly 200 US Occupy campaigns. Another project team uses TextThresher to assess news articles, finding evidence of misinformation and disinformation. Dr. Adams has built research communities across the Berkeley campus, founding D-Lab’s Computational Text Analysis Working Group, BIDS' Text Across Domains (Text XD) initiative, and a set of Data Discovery Research projects converting digital textual archives into research-ready databases queryable by computational text analysts and the public. Look for his book in 2019: Hybrid Text Analysis: Humans and Machines Together.