Instructions New Game More Free Stuff

Tuesday, April 10, 2012

MITH Digital Dialogue: Jordan Boyd-Graber, "Making Topic Models More Human(e)"

Imagine you need to get the gist of whats going on in a large text dataset such as all tweets that mention Obama, all e-mails sent within a company, or all newspaper articles published by The New York Times in the 1990s. Topic models, which automatically discover the themes which permeate a corpus, are a popular tool for discovering whats being discussed. However, topic models arent perfect; errors hamper adoption of the model, performance in downstream computational tasks, and human understanding of the data. However, humans can easily diagnose and fix these errors. We present a statistically sound model to incorporate hints and suggestions from humans to iteratively refine topic models to better model large datasets. Jordan Boyd-Graber is an assistant professor in Marylands iSchool and UMIACS, and a member of the Cloud Computing Center and the Computational Linguistics and Information Processing (CLIP) Lab. {Will appear in FYI on Apr 9, 2012
Start Time:
12:30 PM
End Time:
1:45 PM
MITH Conference Room
Common Location Name:
McKeldin Library
Web Address:
Other Contact Information:
Emma M Millon +1 301 405 9887

Copyright 2012 University of Maryland | Privacy
Contact us with comments, questions and feedback