Learning to merge search results for efficient Distributed Information Retrieval

Kien Tjin-Kam-Jet, Djoerd Hiemstra

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

19 Downloads (Pure)

Abstract

Merging search results from different servers is a major problem in Distributed Information Retrieval. We used Regression-SVM and Ranking-SVM which would learn a function that merges results based on information that is readily available: i.e. the ranks, titles, summaries and URLs contained in the results pages. By not downloading additional information, such as the full document, we decrease bandwidth usage. CORI and Round Robin merging were used as our baselines; surprisingly, our results show that the SVM-methods do not improve over those baselines.
Original languageUndefined
Title of host publicationThe 10th Dutch-Belgian Information Retrieval Workshop
Place of PublicationNijmegen
PublisherRadboud University
Pages55-62
Number of pages8
ISBN (Print)not assigned
Publication statusPublished - 25 Jan 2010
Event10th Dutch-Belgian Information Retrieval Workshop, DIR 2010 - Nijmegen, Netherlands
Duration: 25 Jan 201025 Jan 2010
Conference number: 10

Publication series

Name
PublisherRadboud University

Conference

Conference10th Dutch-Belgian Information Retrieval Workshop, DIR 2010
Abbreviated titleDIR
CountryNetherlands
CityNijmegen
Period25/01/1025/01/10

Keywords

  • Distributed Information Retrieval
  • EWI-17664
  • interleaving
  • collection fusion
  • learning to rank
  • meta-search
  • results merging
  • IR-70344
  • METIS-270757
  • DB-IRNOX: INFORMATION RETRIEVAL (NON-XML)
  • Federated search
  • round robin

Cite this

Tjin-Kam-Jet, K., & Hiemstra, D. (2010). Learning to merge search results for efficient Distributed Information Retrieval. In The 10th Dutch-Belgian Information Retrieval Workshop (pp. 55-62). Nijmegen: Radboud University.