Effective query expansion for federated search

Milad Shokouhi*, Leif Azzopardi, Paul Thomas

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

20 Citations (Scopus)

Abstract

While query expansion techniques have been shown to improve retrieval performance in a centralized setting, they have not been well studied in a federated setting. In this paper, we consider how query expansion may be adapted to federated environments and propose several new methods: where focused expansions are used in a selective fashion to produce specific queries for each source (or a set of sources). On a number of different testbeds, we show that focused query expansion can significantly outperform the previously proposed global expansion method, and - -contrary to earlier work - -show that query expansion can improve performance over standard federated retrieval. These findings motivate further research examining the different methods for query expansion, and other forms of system and user interaction, in order to continue improving the performance of interactive federated search systems.

Original languageEnglish
Title of host publicationProceedings - 32nd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2009
Pages427-434
Number of pages8
DOIs
Publication statusPublished - 2009
Externally publishedYes
Event32nd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2009 - Boston, MA, United States
Duration: 19 Jul 200923 Jul 2009

Publication series

NameProceedings - 32nd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2009

Conference

Conference32nd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2009
Country/TerritoryUnited States
CityBoston, MA
Period19/07/0923/07/09

Fingerprint

Dive into the research topics of 'Effective query expansion for federated search'. Together they form a unique fingerprint.

Cite this