Twenty-One at TREC-7: ad-hoc and cross-language track

Djoerd Hiemstra, Wessel Kraaij

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

113 Downloads (Pure)


This paper describes the official runs of the Twenty-One group for TREC-7. The Twenty-One group participated in the ad-hoc and the cross-language track and made the following accomplishments: We developed a new weighting algorithm, which outperforms the popular Cornell version of BM25 on the ad-hoc collection. For the CLIR task we developed a fuzzy matching algorithm to recover from missing translations and spelling variants of proper names. Also for CLIR we investigated translation strategies that make extensive use of information from our dictionaries by identifying preferred translations, main translations and synonym translations, by defining weights of possible translations and by experimenting with probabilistic boolean matching strategies.
Original languageUndefined
Title of host publicationProceedings of the seventh Text Retrieval Conference (TREC)
EditorsE.M Voorhees, D.K. Harman
Place of PublicationGaithersburg, USA
PublisherNational Institute of Standards and Technology
Number of pages12
Publication statusPublished - 1999
EventSeventh Text REtrieval Conference, TREC-7 1998 - Gaithersburg, United States
Duration: 9 Nov 199911 Nov 1999
Conference number: 7

Publication series

NameNIST Special Publications
PublisherUS National Institute of Standards and Technology


ConferenceSeventh Text REtrieval Conference, TREC-7 1998
Abbreviated titleTREC
Country/TerritoryUnited States


  • CR-H.3.3
  • IR-66980
  • EWI-9421
  • METIS-119693

Cite this