Natural language research announcements

["Fleck, Margaret M" <mfleck AT cs.uiuc.edu>]

________________________________________
From: 
dais-bounces AT cs.uiuc.edu
 
[dais-bounces AT cs.uiuc.edu]
 On Behalf Of Zhai, Chengxiang 
[czhai AT cs.uiuc.edu]
Sent: Wednesday, June 17, 2009 10:03 PM
To: 
dais AT cs.uiuc.edu
Subject: [dais] Special IR/NLP talk: Ken Church, Friday, June 19,       
10:30-11:30am

There will be a special Information Retrieval/Natural Language Processing 
talk this Friday, June 19 at 10:30-11:30am to be given by  Prof. Ken Church 
from Johns Hopkins University. See below for details.

Ken has done a lot of influential work in the general area of human language 
technologies. You may find more information about his work from his online CV:

http://research.microsoft.com/en-us/um/people/church/wwwfiles/resume.html

This page isn't entirely up to date as he has recently moved to Johns Hopkins 
University.

Hope to see many of you at the seminar!

-Cheng

============ Special IR/NLP Talk Announcement =============

Title: Repetition and Language Models
Speaker: Ken Church, Johns Hopkins University
Time: 10:30-11:30am, Friday, June 19, 2009
Place: 3401 Siebel Center
Abstract:
Repetition is very common. Standard bag-of-word models in Information 
Retrieval do not attempt to model discourse structure such as given/new. The 
first mention in a news article (e.g., "Manuel Noriega, former President of 
Panama") is different from subsequent mentions (e.g., "Noriega"). Adaptive 
language models were introduced in Speech Recognition to capture the fact 
that probabilities change or adapt. After we see the first mention, we should 
expect a subsequent mention. If the first mention has probability p, then 
under standard (bag-of words) independence assumptions, two mentions ought to 
have probability p^2, but we find the probability is actually closer to p/2. 
Adaptation matters more for meaningful units of text. In Japanese, words 
(meaningful sequences of characters) are more likely to be repeated than 
fragments (meaningless sequences of characters from words that happen to be 
adjacent). In newswire, we find more adaptation for content words (proper 
nouns, technical terminology and good keywords for information retrieval), 
and less adaptation for function words, clichés and ordinary first names. 
There is more to meaning than frequency. Content words are not only low 
frequency, but likely to be repeated.
Bio:
Kenneth Church is Chief Scientist of the Human Language Technology Center of 
Excellence at the Johns Hopkins University. Before taking this position, he 
was a lead researcher at Microsoft Research in Redmond, and before that he 
was the head of a data mining department in AT&T Labs-Research (formally AT&T 
Bell Labs) where he was named
an AT&T Fellow in 2001.  He received his BS, Masters and PhD from  MIT in 
computer science in 1978, 1980 and 1983, respectively. He enjoys applying 
scalable techniques to very large corpora to address problems in many areas 
of computational linguistics. He has worked and published on many  topics 
including: web search, language modeling, text analysis, spelling correction, 
word-sense disambiguation, terminology, translation, lexicography, 
compression, speech (recognition and synthesis), OCR, as well as applications 
 that go well beyond computational linguistics such as revenue  assurance and 
virtual integration (using screen scraping and web  crawling to integrate 
systems that traditionally don't talk together as well as they could such as 
billing and customer care).



_______________________________________________
dais mailing list
dais AT cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/dais