Forum for AI: Dr. Ed Hovy/USC Information Sciences Institute Learning by Reading: An Experiment in Text Analysis in ACES 2.302

Contact Name: 
Jenna Whitney
Date: 
Apr 14, 2006 11:00am - 12:00pm


There is a signup schedule for this event.

Speaker
Name/Affiliation: Dr. Ed Hovy/USC Information Sciences Institute

T

alk Title: Learning by Reading: An Experiment in Text Analysis

Date

/Time: April 14 2006 at 11:00 a.m.

Location: ACES 2.302

Ho

st: Dr. Bruce Porter

Talk Abstract:
A few years ago three resea

rch groups participated in an
audacious experiment called Project Halo:
(manually) converting
the information contained in one chapter of a hi

gh school
chemistry textbook into knowledge representation statements

and then having the knowledge representation system take
the high s

chool AP exam. Surprisingly all three systems
passed albeit at a rel

atively low level of performance.
Could one do the same automatically

? If not fully how
far can one go? Since October several projects ha

ve taken
up this challenge or aspects of it. Our Learning by Reading

project at ISI drawing part-time participation of experts
in NLP a

nd KR&R addresses the problem from the perspective
of NLP. After suit

able analysis and preparation we parse
the Chemistry textbook and then
convert the results into
very shallow pre-logic predications which ar

e asserted
to a knowledge base. The evaluation still in progress
has two aspects. In the first we apply questions to the
system at va

rious levels (text-only knowledge level without
inference the latter<

br>with inference) and compare performance. In the second
we (in con

junction with other groups) compare the systems
bottom-up automatically
derived representations to the top-down
ones created by hand by those

groups. Although this project
(and related) projects are merely pilot

studies they nonetheless
are likely to generate some interesting concl

usions regarding
the gap between what automated systems can deliver and
what
human knowledge engineers deem necessary in the fascinating

endeavor of learning by reading.

Speaker Bio:
Eduard Hovy leads

the Natural Language Research Group at
the Information Sciences Institu

te of the University of
Southern California. He is also Deputy Directo

r of the
Intelligent Systems Division as well as a research associate

professor of the Computer Science Department of USC and
Advisory Pr

ofessor of the Beijing University of Posts and
Telecommunications. He c

ompleted a Ph.D. in Computer Science
(Artificial Intelligence) at Yale

University in 1987. His
research focuses on information extraction au

tomated text
summarization question answering the semi-automated cons

truction
of large lexicons and ontologies machine translation and
digital government. Dr Hovy regularly serves in an advisory
capacity

to funders of NLP research in the US and EU. He
is the author or co-ed

itor of five books and over 170 technical
articles. In 2001
Dr. Hov

y served as President of the Association for Computational
Linguistics

(ACL) and in 2001-03 as President of the International
Association of M

achine Translation (IAMT). Dr. Hovy regularly
co-teaches a course in th

e Masters Degree Program in Computational
Linguistics at the University
of Southern California as
well as occasional short courses on MT and

other topics
at universities and conferences. He has served on the Ph.D. and M.S. committees for students from USC Carnegie
Mellon Univ

ersity the Universities of Toronto Karlsruhe
Pennsylvania Stockholm
Waterloo Nijmegen Pretoria and
Ho Chi Minh City.