Title: Thinking outside the box for Natural Language Processing
Location: Graduate Center Science Center Room 4102, 365 Fifth Ave., New York, NY
Time: 2:15-3:30
Date: Nov 4, 2011


Thinking outside the box for Natural Language Processing

            Srinivas Bangalore
            AT&T Labs-Research
          Florham Park, NJ 07932
          srini@research.att.com

Abstract: Natural Language Processing (NLP) has traditionally been
thought of as a
sequence of transformation boxes with each box progressively
adding  richer information to the input. While this block structure
affords modularity in design of NLP
systems, it is prone to error propagation from one step to the next.

In this talk, I will present three instances of incorporating richer
information derived from "outside the box", in the context of
Speech-to-speech translation, Speech driven Question-Answering and
modeling human-human dialogs tasks.


Bio:

Dr. Srinivas Bangalore is currently a Principal Member of Technical
Specialist at AT&T Labs-Research. He received his PhD in Computer
Science from University of Pennsylvania. Since joining AT&T in 1997,
he has worked on many areas of natural language processing including
Spoken Language Translation, Multimodal Understanding, Language
Generation and Question-Answering. He has co-edited a book on
Supertagging, authored over a 100 research publications and holds over
30 patents in these areas. Dr. Bangalore has been adjunct associate
professor at Columbia University and a visiting lecturer at Pricenton
University. He has been awarded the AT&T Outstanding Mentor Award and
the AT&T Science & Technology Medal.  He has served as an editorial
board member of Computational Linguistics Journal and a program
committee member for a number of ACL and IEEE Speech conferences.