University of Twente at the TREC 2007 Enterprise Track : Modeling relevance propagation for the expert search task

Pavel Serdyukov, H. Rode, Djoerd Hiemstra

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademic

66 Downloads (Pure)


This paper describes several approaches which we used for the expert search task of the TREC 2007 Enterprise track. We studied several methods of relevance propagation from documents to related candidate experts. Instead of one- step propagation from documents to directly related candi- dates, used by many systems in the previous years, we do not limit the relevance ow and disseminate it further through mutual documents-candidates connections. We model rele- vance propagation using random walk principles, or in for- mal terms, discrete Markov processes. We experiment with innite and nite number of propagation steps. We also demonstrate how additional information, namely hyperlinks among documents, organizational structure of the enterprise and relevance feedback may be utilized by the presented techniques.
Original languageUndefined
Title of host publicationProceedings of the 16th Text Retrieval Conference (TREC)
EditorsE.M Voorhees, Lori P. Buckland
Place of PublicationGaithersburg, USA
PublisherNIST Special Publication
Number of pages5
ISBN (Print)not assigned
Publication statusPublished - 15 Feb 2008
Event16th Text REtrieval Conference, TREC 2007 - Gaithersburg, United States
Duration: 6 Nov 20079 Nov 2007
Conference number: 16

Publication series

NameNIST Special Publication
PublisherNational Institute of Standards and Technology (NIST)
VolumeSP 500-274


Conference16th Text REtrieval Conference, TREC 2007
Abbreviated titleTREC
Country/TerritoryUnited States


  • EWI-12314
  • METIS-250967
  • IR-64732

Cite this