Skip to main content

Research Repository

Advanced Search

Robust query-specific pseudo feedback document selection for query expansion.

Huang, Qiang; Song, Dawei; R�ger, Stefan


Qiang Huang

Dawei Song

Stefan R�ger


Craig Macdonald

Iadh Ounis

Vassilis Plachouras

Ian Ruthven

Ryen W. White


In document retrieval using pseudo relevance feedback, after initial ranking, a fixed number of top-ranked documents are selected as feedback to build a new expansion query model. However, very little attention has been paid to an intuitive but critical fact that the retrieval performance for different queries is sensitive to the selection of different numbers of feedback documents. In this paper, we explore two approaches to incorporate the factor of query-specific feedback document selection in an automatic way. The first is to determine the \optimal number of feedback documents with respect to a query by adopting the clarity score and cumulative gain. The other approach is that, instead of capturing the optimal number, we hope to weaken the effect of the numbers of feedback document, i.e., to improve the robustness of the pseudo relevance feedback process, by a mixture model. Our experimental results show that both approaches improve the overall retrieval performance.


HUANG, Q., SONG, D. and RUGER, S. 2008. Robust query-specific pseudo feedback document selection for query expansion. In Macdonald, C., Ounis, I., Plachouras, V., Ruthven I. and White, R.W. (eds.) Advances in information retrieval: proceedings of the 30th European conference on information retrieval (IR) research (ECIR 2008), 30 March - 3 April 2008, Glasgow, UK. Lecture notes in computer science, 4956. Berlin: Springer [online], pages 547-554. Available from:

Conference Name 30th European conference on information retrieval (IR) research (ECIR 2008)
Conference Location Glasgow, UK
Start Date Mar 30, 2008
End Date Apr 3, 2008
Acceptance Date Dec 31, 2008
Online Publication Date Dec 31, 2008
Publication Date Dec 31, 2008
Deposit Date May 29, 2009
Publicly Available Date May 29, 2009
Publisher Springer
Pages 547-554
Series Title Lecture notes in computer science
Series Number 4956
ISBN 3540786457; 9783540786450
Keywords Document retrieval; Query expansion
Public URL


You might also like

Downloadable Citations