Natural language research announcements

[Ming-Wei Chang <mchang21 AT uiuc.edu>]

Dear faculty and students,

Due to some conflicts, we have to move the talk of Dr. Douglas Downey
(details below) earlier.  His talk will now start at 3:15pm and end at
4:15pm, on Feb 19th (this Thursday). The room number is 3405. Hope to
see you there!

Thank you,
Ming-Wei


> Title: Autonomous Web-scale Information Extraction
>
> Abstract:
> Search engines are extremely useful tools for answering questions.  However,
> a significant number of questions users might pose -- for example, "which
> nanotechnology companies are hiring on the West Coast?" -- cannot be
> addressed using existing search engines, because the answers do not lie on a
> single page.  To answer these kinds of queries, users must extract and
> synthesize information from multiple documents.  Currently, this is a
> tedious and error-prone manual process.
>
> In this talk, I will describe my research aimed at automating the extraction
> of this information from the Web.  I will present a model of the redundancy
> inherent in the Web, and show that the model can be used to identify correct
> extractions autonomously, without the manually labeled examples typically
> assumed in previous information extraction research.  However, the model has
> limited efficacy for the "long tail" of infrequently mentioned facts; I
> demonstrate how unsupervised language models can be leveraged in concert
> with redundancy to overcome this limitation.  Lastly, I will describe recent
> theoretical and experimental results illustrating that a generalization of
> the redundancy-based approach is effective for a variety of textual
> classification tasks, beyond information extraction.
>
> Bio:
> Doug Downey is an assistant professor in the EECS Department of Northwestern
> University, which he joined in the Fall of 2008.  He obtained his PhD from
> the University of Washington, where he was advised by Oren Etzioni and
> supported by an NSF Fellowship and Microsoft Research Graduate Fellowship.
> His research interests are in the areas of natural language processing,
> machine learning, and artificial intelligence.  At UW, he was part of the
> KnowItAll project, which was aimed at utilizing the Web to autonomously
> extract large knowledge bases.  Doug's primary research results concern
> probabilistic models of the redundancy inherent in large corpora, along with
> associated techniques that allow systems like KnowItAll to extract data
> autonomously at high precision.