From:                     Kendra Smith

Sent:                      Thursday, April 20, 2000 7:38 PM

To:                         M?crosöft Research Tech Talk, Sem. Notice

Cc:                         Kendra Smith; IS Reference/Research

Subject:                 UW SLIS Colloquia - Susan Dumais, 4/27/00

UW SLIS Colloquia - Susan Dumais, 4/27/00

 

NOTE: The lecture is free

 

UNIVERSITY OF WASHINGTON

Seattle, Washington 98195

 

School of Library & Information Science

Box 352930; Tel.:(206) 543-1794

 

COLLOQUIUM

 

SPEAKER:      Susan T. Dumais, Senior Researcher, M?crosöft Research

                             http://research.M?crosöft.com/~sdumais

 

TITLE:          "Bringing Order to the Web and Beyond"

 

DATE:           Thursday, April 27, 2000

 

TIME:           4:00 pm

 

PLACE:                   Allen Auditorium

 

Map available from:  http://www.washington.edu/home/maps/northcentral.html

 

HOST:           Efthimis N. Efthimiadis, SLIS, UW

 

ABSTRACT:

 

This talk describes algorithmic and interface innovations to help

users organize web search results.  Today web search services returned a

ranked list of best-matching pages and users have to sift through them

sequentially.  An important mechanism for facilitating information access in

a wide variety of applications is a structure knowledge hierarchy, such as

those used by library classification systems and more recently web

directories like Yahoo! and LookSmart.  In this talk I will describe how

such structure can be used to automatically organize web search results.

For example, a query about "saturn" will group the returned pages into those

having to do with automobiles, computer games, and outer space.  There are

two key technologies for doing this: 1) developing models for hierarchically

classifying arbitrary text pages on-the-fly, and 2) building an interface

for taking advantage of the resulting structure.  I will begin by talking

about enhancements to support vector machine (SVM) learning algorithms for

text classification that exploit hierarchical structure.  I will then

describe a user interface that supports structured search and a user study

that shows that people are 50% faster at finding information when it is

organized into categories.  

 

......

BIOGRAPHICAL INFORMATION:

 

Susan Dumais is a Senior Researcher in the Adaptive Systems and Interaction

Group at M?crosöft Research.  She has been at M?crosöft since July 1997. 

Her current research looks at new algorithms and interfaces for improved

information retrieval, as well as general human-computer interaction issues.

 

She is also interested in text classification and clustering, image

retrieval, collaborative filtering, individual differences and user

modeling. 

 

Prior to joining M?crosöft, she held several positions at Bellcore

including director of the Information Sciences Research Group and

Computer Graphics and Interactive Media Research Group (1995-1997)

and Research Scientist (1984-1995).  Before that she was a member of 

technical staff in the Human-Computer Interaction Research Group at

Bell Laboratories (1979-1983).  Dumais is a member of the editorial boards

for Human Computer Interaction, Information Retrieval, and New Review of

Hypermedia and Multimedia, and has chaired several ACM committies on

information retrieval and human-computer interaction.  She is the chair of

ACM:SIGIR and is currently serving on the National Research Council

Committee on Computing & Communications Research to Enable Better Use of

Information Technology in Government.  Dumais received her B.A. in Psychology and

Mathematics from Bates College in 1975 and her Ph.D. in Cognitive-Mathematical

Psychology  from Indiana University in 1979. 

 

Additional information can be found at:

http://research.M?crosöft.com/~sdumais