A probabilistic model based approach for blended search

ABSTRACT

In this paper, we propose to model the blended search problem by assuming conditional dependencies among queries, VSEs and search results. The probability distributions of this model are learned from search engine query log through unigram language model. Our experimental exploration shows that, (1) a large number of queries in generic Web search have vertical search intentions; and (2) our proposed algorithm can effectively blend vertical search results into generic Web search, which can improve the Mean Average Precision (MAP) by as much as 16% compared to traditional Web search without blending.

REFERENCES

• 1: Borthwick, A. "Survey Paper on Statistical Language Modeling", Technical Report, Proteus project, New York University Computer Science Department, 1997
• 2: Gauch, S., Wang, G. and Gomez, M. Profusion: intelligent fusion from multiple, distributed search engines. J. Univers. Comput. Sci. 2(9), (1996), 637--649.
• 3: Kalervo Järvelin , Jaana Kekäläinen, IR evaluation methods for retrieving highly relevant documents, Proceedings of the 23rd annual international ACM SIGIR conference on Research and development in information retrieval, p.41-48, July 24-28, 2000, Athens, Greece [doi>10.1145/345508.345545]
• 4: Weiyi Meng , Clement Yu , King-Lup Liu, Building efficient and effective metasearch engines, ACM Computing Surveys (CSUR), v.34 n.1, p.48-89, March 2002 [doi>10.1145/505282.505284]

INDEX TERMS

Primary Classification:
  H. Information Systems
  H.3 INFORMATION STORAGE AND RETRIEVAL
      H.3.3 Information Search and Retrieval

General Terms:
Algorithms, Verification

Keywords:
blended search, language model, query log, vertical search