Computing Reviews
Today's Issue Hot Topics Search Browse Recommended My Account Log In
Review Help
Search
Challenges in information retrieval and language modeling: report of a workshop held at the center for intelligent information retrieval, University of Massachusetts Amherst, September 2002
Allan J., Aslam J., Belkin N., Buckley C., Callan J., Croft B., Dumais S., Fuhr N., Harman D., Harper D., Hiemstra D., Hofmann T., Hovy E., Kraaij W., Lafferty J., Lavrenko V., Lewis D., Liddy L., Manmatha R., McCallum A., Ponte J., Prager J., Radev D., Resnik P., Robertson S., Rosenfeld R., Roukos S., Sanderson M., Schwartz R., Singhal A., Smeaton A., Turtle H., Voorhees E., Weischedel R., Xu J., Zhai C. ACM SIGIR Forum37 (1):31-47,2003.Type:Article
Date Reviewed: Jan 23 2004

In the last ten years, information retrieval (IR) has evolved from a niche field into an important and multifaceted discipline, and has produced measurable results that affect the daily life of millions. No less important, its theoretical foundations have been substantially advanced by a new research paradigm based on language modeling (LM).

In September 2002, a group of leading researchers in IR and LM gathered for a two-day workshop with the goal of discussing the current status of IR and answering some basic questions: Where are we? Where do we need to go? How can LM help? This paper is a final report from that workshop, edited by Allan and Croft, and taken from the contributions of all 36 participating researchers. It consists of two main sections: one on long-term challenges, and one on short-term issues.

In the first section, two main grand challenges for IR are identified: global information access, namely, satisfying complex and diverse types information needs by integrating diverse sources of structured and unstructured data; and contextual retrieval, namely, leveraging knowledge of the user context in order to always provide the right results in the right context. In the second section, 11 major subfields of IR are analyzed: retrieval models; cross-lingual retrieval; Web search; user modeling for IR; filtering, topic detection and tracking, and classification; summarization; question answering; metasearch and distributed retrieval; multimedia retrieval; information extraction; and testbeds. Each is discussed by first briefly introducing what the subfield is about, then by describing its short-term challenges, highlighting the resources needed to meet these challenges, and discussing how LM can help.

The report is very well written and well structured. Unusually for an edited work, it is also very even in its treatment of the various subfields of IR, each of which gets the same amount of space, and each of which is discussed in clear prose, at the same level of granularity. This paper will be of great interest to all IR researchers, and in general, to anyone wishing to know where IR is and where its next research avenues are going to lead. Above all, this report is a resource of exceptional value for policy makers concerned with the development of the global information society in the next several decades.

Reviewer:  F. Sebastiani Review #: CR128986 (0407-0845)
Bookmark and Share
 
Information Search And Retrieval (H.3.3 )
 
 
Language Models (I.2.7 ... )
 
Would you recommend this review?
yes
no
Other reviews under "Information Search And Retrieval": Date
Nested transactions in a combined IRS-DBMS architecture
Schek H. (ed)  Research and development in information retrieval (, King’s College, Cambridge,701984. Type: Proceedings
Nov 1 1985
An integrated fact/document information system for office automation
Ozkarahan E., Can F. (ed) Information Technology Research Development Applications 3(3): 142-156, 1984. Type: Article
Oct 1 1985
Access methods for text
Faloutsos C. ACM Computing Surveys 17(1): 49-74, 1985. Type: Article
Jan 1 1986
more...

E-Mail This Printer-Friendly
Send Your Comments
Contact Us
Reproduction in whole or in part without permission is prohibited.   Copyright 1999-2024 ThinkLoud®
Terms of Use
| Privacy Policy