Learning and Reasoning over Social, Behavioral and Textual Information

Speaker Name: 
Dan Goldwasser
Speaker Title: 
Assistant Professor of Computer Science
Speaker Organization: 
Purdue University
Start Time: 
Thursday, May 24, 2018 - 11:45am
End Time: 
Thursday, May 24, 2018 - 1:15pm
UCSC, Engineering 2, Room 192
Lise Getoor & Lyn Walker

Abstract: Natural language provides the means for human communication, allowing us to exchange ideas and interact on a wide range of issues. However, when it comes to computationally modeling these interactions, we typically ignore their broader context and analyze the text in isolation. In my talk I will review several recent works which demonstrate the importance of holistically modeling behavioral, social and textual information. We will focus on two main applications, political discourse analysis on Twitter and debate discussions.

Bio: Dan Goldwasser is an assistant professor in the Department of Computer Science at Purdue University. He graduate from the University of Illinois at Urbana-Champaign at 2012 and has spent two years at the University of Maryland as a postdoctoral researcher. His research interests are in the area of natural language processing and applications of machine learning for textual data. His work focuses on real world applications, such as machine-comprehension and political discourse analysis. These applications require machine learning algorithms capable of connecting the raw textual data with world-knowledge and dynamic real-world behaviors.