Seminar by Prof. Sudeshna Sarkar
Use of Domain Knowledge and User Profile for Personalized Content Retrieval and Presentation
Prof. Sudeshna Sarkar
Computer Science and Engineering Department
Indian Institute of Technology Kharagpur
Date: Thursday, September 26, 2002
Time: 04:15 PM
Venue: CS-101
Abstract
The Internet is touted to provide a wealth of information instantaneously and on demand. However, because of the sheer quantity of information available, it is extremely important to be able to retrieve quality information relevant to a given user. Many powerful search engines exist that index a large useful portion of the Internet, but the information retrieved by the search engine is user independent, and is not able to take care of the personalized requirements, interest and level of understanding of individual users. We look at how we can use domain knowledge of a topic, in conjunction with a representation of the user profile to decide the relevance of a document to the user. Our system also creates a personalized presentation and packaging of a Web site given an user's profile. It also aims to build up an internal repository of documents and links of sites most popular among a group of users.
This system has been designed to be of value to users in rural schools in India. These schools cannot afford a persistent Internet connection, contain a homogeneous group of users of similar level, and similar socio-economic background, and hence share a lot of common interest. The system will help the schools to consume minimum bandwidth and be able to obtain information of value.
About the Speaker
Sudeshna Sarkar is currently an Associate Professor in the Department of Computer Science and Engineering at IIT Kharagpur. She completed her B.Tech. in 1989 from IIT Kharagpur, MS from University of California, Berkeley, and PhD from IIT Kharagpur in 1995. She has been in the faculty of IIT Guwahati and at IIT Kanpur before joining IIT Kharagpur. Her research interests are in Artificial Intelligence. She has been working on search and planning, in machine learning, and in information retrieval and natural language processing tools for Bengali and Hindi. She has been the Prioncipal Investigaor of several projects sponsored by AICTE, Tata Infotech Ltd, MHRD, and Media Lab Asia.