Natural language research announcements

["Alexandre Klementiev" <klementi AT uiuc.edu>]

Dear faculty and students,

Dr. Ivan Titov will give a talk (details below) at the AIIS
 seminar this Thursday.

Thank you,


Alex



Title: Modeling Online Reviews: Exploiting User Annotations for Sentiment Summarization


Speaker: 

Ivan Titov, UIUC




Date: Mar. 27, 4:00pm
Location:    Siebel 3405






Abstract:

User generated content represents a unique source of information in which user interface tools have facilitated the creation of an abundance of labeled content, e.g., topics in blogs, numerical product and service ratings in user reviews, and helpfulness rankings in online discussion forums. Many previous studies on user generated content have attempted to predict these labels automatically from the associated text. However, these annotations are often present in the data already, which opens another interesting line of research: designing models leveraging these labelings to improve a wide variety of applications. In this talk I will be considering the sentiment summarization problem. I will present statistical models which exploit user generated aspect ratings to discover corresponding topics and are therefore able to extract fragments of text discussing these aspects without the need of annotated data. Joint work with Ryan McDonald.

Bio:

Ivan Titov is a postdoctoral researcher at the Cognitive Computation Group working with Dan Roth. He has just received his PhD degree in computer science from the University of Geneva, Switzerland with the thesis focused on latent variable models for natural language parsing.  He has started working on topic modeling and sentiment summarization during his internship at Google Research in Summer, 2007. His research interests include statistical natural language processing, machine learning for structure processing, kernel methods and graphical models.