Skip to main navigation Skip to search Skip to main content

A relevance-novelty combined model for genomics search result diversification

  • Xiaoshi Yin*
  • , Zhoujun Li
  • , Jimmy Xiangji Huang
  • , Xiaohua Hu
  • *Corresponding author for this work
  • Beihang University
  • York University Toronto
  • Drexel University

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

Abstract

Traditional retrieval models assume that the relevance of a document is independent of the relevance of other documents. However, this assumption may result in high redundancy and low diversity in a ranked list. In order to provide comprehensive and diverse answers to fulfill biologists' information need, we propose a relevance-novelty combined model, named RelNov model, based on the framework of an undirected graphical model. Experiments conducted on the TREC 2006 and 2007 Genomics collections show that the proposed approach is effective in promoting both diversity and relevance of retrieval ranked lists.

Original languageEnglish
Title of host publicationProceedings - 2010 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2010
Pages692-695
Number of pages4
DOIs
StatePublished - 2010
Event2010 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2010 - Hong Kong, China
Duration: 18 Dec 201021 Dec 2010

Publication series

NameProceedings - 2010 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2010

Conference

Conference2010 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2010
Country/TerritoryChina
CityHong Kong
Period18/12/1021/12/10

Keywords

  • Diversity
  • Genomics search
  • Graphical model

Fingerprint

Dive into the research topics of 'A relevance-novelty combined model for genomics search result diversification'. Together they form a unique fingerprint.

Cite this