UTCS Colloquium/AI-Noah Smith/CMU: "Text-Driven Forecasting: Meaning as a Real Number," TAY 3.128, Friday, April 9, 2010, 11:00 a.m.

Contact Name: 
Jenna Whitney
Date: 
Apr 9, 2010 11:00am - 12:00pm

There is a sign-up schedule for this event that can be found
at http://www.cs.utexas.edu/department/webeven

t/utcs/events/cgi/list_events.cgi

Type of Talk: UTCS Colloquium/A

I

Speaker/Affiliation: Noah Smith/CMU

Date/Time: Friday, Apri

l 9, 2010, 11:00 a.m.

Location: TAY 3.128

Host: Ray Mooney

Talk Title: Text-Driven Forecasting: Meaning as a Real Number

Ta

lk Abstract:

We take inspiration from recent research on sentiment an

alysis that
interprets text based on the subjective attitude of the au

thor. We consider
related tasks where a piece of text is interpreted t

o predict some
extrinsic, real-valued outcome of interest that can be
observed in non-text
data. Examples include:

The interpret

ation of an annual financial report from a company to its
shareholders
is the risk incurred by investing in the company in the coming
year.<

br />
* The interpretation of a critic''s review of a film is the film

''s box office
success.
* The interpretation of a political blog

post is the response it garners
from readers.
* The interpretatio

n of a day''s microblog feeds is the public''s opinion
about a particu

lar issue.

In all of these cases, one aspect of the text''s meaning

is observable from
objective real-world data, although perhaps not im

mediately at the time the
text is published (respectively: return vola

tility, gross revenue, user
comments, and traditional polls). We pr

opose a generic approach to
text-driven forecasting that is expected t

o benefit from linguistic analysis
while remaining neutral to differen

t theories of language. A highly
attractive property of this line of r

esearch is that evaluation is
objective, inexpensive, and theory-neu

tral. This approach introduces some
methodological challenges, as wel

l.
We conjecture that forecasting tasks, when considered in concert,
will be a
driving force in domain-specific, empirical, and extrinsi

cally useful
natural language analysis. Further, this research direct

ion will push NLP to
consider the language of a more diverse subset of
the population, and may
support inquiry in the social sciences about
foreknowledge and communication
in societies.

This talk in

cludes joint work with Ramnath Balasubramanyan, William Cohen,
Dipan

jan Das, Kevin Gimpel, Mahesh Joshi, Shimon Kogan, Dimitry Levin,
Brendan O''Connor, Bryan Routledge, Jacob Sagi, and Tae Yano.

Sp

eaker Bio:

Noah Smith is an assistant professor in the School of Comp

uter Science at
Carnegie Mellon University. He received his Ph.D. in C

omputer Science, as a
Hertz Foundation Fellow, from Johns Hopkins Un

iversity in 2006 and his B.S.
in Computer Science and B.A. in Linguist

ics from the University of Maryland
in 2001. His research interests in

clude statistical natural language
processing, especially unsupervise

d methods, machine learning for structured
data, and applications of
natural language processing. He serves on the
editorial board of the

journal Computational Linguistics and received a best
paper award at t

he ACL 2009 conference. His ten-person group, Noah''s ARK, is
suppor

ted by the NSF, DARPA, Qatar NRF, Portugal FCT, and gifts from Google,

HP Labs, and IBM Research.