The Co-Effects of Query Structure and Expansion on RetrievalPerformance in Probabilistic Text Retrieval

作者: Jaana Kekäläinen , Kalervo Järvelin

DOI: 10.1023/A:1009983401464

关键词:

摘要: The effects of query structures and expansion (QE) on retrieval performance were tested with a best match system (InQuery^1). Query structure means the use operators to express relations between search keys. Six different tested, representing strong (e.g., queries facets or concepts identified) weak (no identified, is ‘a bag keys’). QE was based concepts, which first selected from searching thesaurus, then expanded by semantic relationships given in thesaurus. levels (a) no expansion, (b) synonym (c) narrower concept (d) an associative (e) cumulative all other expansions. With Boolean structured queries, not very effective. achieved combination facet structure, where keys within treated as instances one key (the SYN operator), largest expansion.

参考文章(24)
David Hawking, Nick Craswell, Paul B. Thistlewaite, ANU/ACSys TREC-6 Experiments. text retrieval conference. pp. 275- 290 ,(1997)
James P. Callan, John Broglio, James Allan, Lisa Ballesteros, W. Bruce Croft, Jinxi Xu, Hongming Shu, INQUERY at TREC-5 text retrieval conference. pp. 119- 132 ,(1996)
Jinxi Xu, W. Bruce Croft, Quary Expansion Using Local and Global Document Analysis international acm sigir conference on research and development in information retrieval. ,vol. 51, pp. 4- 11 ,(1996) , 10.1145/3130348.3130364
Chris Buckley, Mandar Mitra, Janet Walz, Claire Cardie, Using clustering and SuperConcepts within SMART: TREC 6 Information Processing & Management. ,vol. 36, pp. 109- 131 ,(2000) , 10.1016/S0306-4573(99)00047-3
Raya Fidel, Efthimis N. Efthimiadis, Terminological knowledge structure for intermediary expert systems Information Processing and Management. ,vol. 31, pp. 15- 27 ,(1995) , 10.1016/0306-4573(94)E0004-L
Kalervo J?rvelin, Jaana Kristensen, Timo Niemi, Eero Sormunen, Hiekki Keskustalo, A deductive data model for query expansion international acm sigir conference on research and development in information retrieval. pp. 235- 243 ,(1996) , 10.1145/243199.248146