Computer Science & EngineeringDepartment
Indian Institute of Technology Kharagpur
WB, India 721302
B.Tech, 1989, IIT Kharagpur
M.S., 1991, University of California, Berkeley
Ph.D., 1996, IIT Kharagpur
Home Page of Sudeshna Sarkar
| Professor
Computer Science & EngineeringDepartment Indian Institute of Technology Kharagpur WB, India 721302 B.Tech, 1989, IIT Kharagpur M.S., 1991, University of California, Berkeley Ph.D., 1996, IIT Kharagpur |
|
Email: shudeshna@gmail.com, sudeshna@cse.iitkgp.ernet.in
Office phone: 03222-283494
Fax: 03222-278985
My broad topics of interest span different areas of artificial
intelligence and machine learning. My current interests are in
text mining and natural language processing, infromation retrieval and data mining.
I am mainly
interested in building large scale practical systems that can be used
to solve real life problems. I wish to make use of natural language
understanding and information extraction methodologies for effective
information retrieval and content recommendation. I am interested in
different forms of text media including noisy text data like speech
transcripts,
transliterated text, SMS and blogs. I also have a special interest in
developing technologies for Indian languages and working with
multilingual issues.
These are some of the topics I have worked on.
Personalized
information retrieval
We have worked on identifying information relevant to the needs of a
given user. We
have worked on acquiring user profile
through direct and indirect methods. We have worked on interfaces to
get the user's interest directly. We have also worked on acquiring user
profiles by tracking the user's sessions and browsing behaviour. We
have worked on systems that capture the
user profile and uses that for identifying relevant documents by
consulting the ontology of the domain. We have also worked on
collaborative filtering based approaches. We have been working
on interfaces for information retrieval systems. We developed a system,
CROWSE that clusters the retrieved results. We have also been working
on appropriate visiualization systems for timed documents like citation
documents.
Natural
Language Technology: I am interested in various aspects of
language technology including natural language processing, use of
language technology in information retrieval, spellcheckers, speech
processing and other areas. We have
developed the following tools for Bengali (বাংলা) and Hindi (हिन्दी). Some of these tools are
available here.