Retrieval based on combining language models with clustering

  • Authors:
  • Hua Huo;Boqin Feng

  • Affiliations:
  • Department of Computer Science, Xi'an, Jiaotong University, Xi'an, P.R. China;Department of Computer Science, Xi'an, Jiaotong University, Xi'an, P.R. China

  • Venue:
  • CIS'04 Proceedings of the First international conference on Computational and Information Science
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

We propose a new retrieval method based on combining language models with clustering. The basic idea of the method is as follows. Firstly, documents in the collection are grouped into clusters by using a clustering algorithm. Secondly, clusters are imported into building language models which are used to estimate how likely a query could be generated from them. Thirdly, language models are smoothed by using a two-stage smoothing method. Our experiments show that the method outperforms both approach “purely” based on clustering and technique “purely” based on language model.