Relevance Ranking

definition: Queries given to search engines or other retrieval systems are often not very specific, and lead to a large number of matching documents. In these cases the retrieval system should have a good estimate of the relevance of the documents to the user's needs, so that "good" documents show up early in the enumeration. A large number of factors should enter into a good ranking method, including the positions of the query terms in the document, linguistic context of the matches, link popularity, classification of the documents, user models etc. "Classical" methods compute a mesaure of "distance" between the query and the retrieved document, such as TF/IDF or cosine similarity. For hyperlinked documents, methods which make use of the hyperlink structure have proved very effective for relevance ranking. Google was the first large-scale search engine to make use of hyperlink sructure for relevance ranking.

related project(s):

Clever

related organisation(s):

Google, Inc.

related person(s):

Jon M. Kleinberg
Gerald Salton
Sergey Brin
Monika R. Henzinger

related system(s) / resource(s):

SMART

relevant source(s):

Wikipedia: Relevance in Information Retrieval
dmoz

related publication(s):

Authoritative sources in a hyperlinked environment.
Jon M. Kleinberg.
Journal of the ACM. 46 (5). 1999. 604--632.
http://citeseer.nj.nec.com/kleinberg97authoritative.html

The anatomy of a large-scale hypertextual Web search engine.
Sergey Brin and Lawrence Page.
Computer Networks and ISDN Systems. 30 (1--7). 1998. 107--117.
http://citeseer.nj.nec.com/brin98anatomy.html

Mining the Web.
Soumen Chakrabarti.
Morgan Kaufmann Publishers. San Francico. 2003.

Link Analysis in Web Information Retrieval.
Monika R. Henzinger.
IEEE Data Engineering Bulletin. 23 (3) 2000. 3-8.

Term-weighting approaches in automatic text retrieval.
Gerard Salton and Christopher Buckley.
Information Processing and Management. 24 (5). 1988. 513-523.

LT World

Relevance Ranking