Title :
PLSA-Based Query Expansion
Author :
Wei-jiang, Li ; Tie-jun, Zhao ; Wen-mao, Zang
Author_Institution :
Comput. Applic. Key Lab. of Yunnan Provice, Kunming Univ. of Sci. & Technol., Kunming, China
Abstract :
The problem of word mismatch is fundamental to information retrieval. The primary goal of this paper is effectively solve the above-mentioned problem by presenting a novel approach to query expansion to LSA and factor analysis, that has a solid statistical foundation, called PLSA, since it is based on the likelihood principle and defines a proper generative model of the data.The experiment result on TREC data collection shows the method proposed by the paper is effective.
Keywords :
probability; query processing; factor analysis; generative model; information retrieval; likelihood principle; probabilistic latent semantic analysis; query expansion; statistical foundation; word mismatch; Buildings; Computer applications; Computer science; Information analysis; Information retrieval; Information science; Laboratories; Probability; Solid modeling; Thesauri; Information Retrieval; PLSA; query expansion;
Conference_Titel :
Computer and Information Science, 2009. ICIS 2009. Eighth IEEE/ACIS International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-0-7695-3641-5
DOI :
10.1109/ICIS.2009.22
