Announcement: UT Turk Users Mailing List
Yinon Bentor
Crowdsourcing, you can ignore this message.]
I'm starting a mailing list for UT people who are using Amazon's
Mechanical Turk, Crowdflower, or other crowdsourcing systems for
research. I hope we can use this list to keep up with each other's
work, announce talks and papers at reading groups that may interest
others using these techniques, and discuss any issues related to IRB
requirements for this type of research.
You can join here:
http://groups.google.com/group/ut-turk-users
If you know of others who may be interested, please forward this
message to them, especially for people outside of CS and LIN.
Thanks,
Yinon
Matt Lease
Date/Time: Friday, August 6, 2010, 3:00 p.m.
Location: UTA 5.522 (in the iSchool, 1616 Guadalupe, across the street
from the Clay Pit)
Host: Matt Lease
Title:
Community-based Question-Answering Systems
Abstract:
Community Question Answering (CQA) services are Internet services that
enable users to ask and answer questions, as well as to search through
historical question-answer pairs. Examples of such services include
Yahoo! Answers, WikiAnswers, etc. CQA services are complementary to
search engines. CQA service Yahoo! Answers had 179 million users and
over 1 billion question and answers as of October 2009.
We aim to improve the current Community Question Answering services. In
my talk, I mainly cover the following topics: 1) Question Search. We
propose to use category information to improve the performance of
question search and develop several new approaches to leveraging
category information. 2) Extracting QA knowledge from online forums. QA
pairs are essential to QA services. Forums contain a huge amount of QA
knowledge, but it is nontrivial to extract the QA knowledge. We develop
some techniques to extract questions and answers from forums. 3) Finding
expert users from a community to answer a new question.
Bio:
Gao Cong is currently an Assistant Professor in the School of Computer
Engineering, Nanyang Technological University (NTU). Before joining NTU,
he was an Assistant professor in Aalborg University, Denmark and worked
as a researcher at the Microsoft Research Asia. From 2004 to 2006, he
worked as a postdoc research fellow in the database group within the
University of Edinburgh. He earned his Ph.D. in Computer Science from
National University of Singapore.
His general research interests include database, data mining, and
information retrieval. His current specific research interests include
search and mining social media, and geo-spatial keyword query
processing. His work was published in database, data mining, and
information retrieval conferences, such as SIGMOD, VLDB, KDD, WWW,
SIGIR, etc.
--
Matt Lease
Assistant Professor
School of Information and Department of Computer Science
University of Texas at Austin
Voice: (512) 471-9350 � Fax: (512) 471-3971 � Office: UTA 5.450
http://www.ischool.utexas.edu/~ml