Title :
The optimal algorithm for query refinement in information retrieval
Author_Institution :
Media Process. Project, NTT Cyberspace Labs., Kanagawa, Japan
Abstract :
To realize more efficient information retrieval it is critical to improve the user´s original query, because novice users can not be expected to formulate precise and effective queries. Queries can often be improved by adding extra terms that appear in relevant documents but which are not included in the original query. This is called query expansion. Query refinement, a variant of query expansion, interactively recommends new terms related to the original query. Since previous research did not offer any criterion to guarantee optimality, this paper proposes an optimal algorithm for query refinement with reference to the Bayes criterion
Keywords :
Bayes methods; Markov processes; iterative methods; learning (artificial intelligence); optimisation; probability; query processing; Bayes criterion; Markov decision process; information retrieval; optimal algorithm; probability; query process; query refinement; reinforcement learning; Decision theory; Feedback; Information retrieval; Laboratories;
Conference_Titel :
Systems, Man, and Cybernetics, 1999. IEEE SMC '99 Conference Proceedings. 1999 IEEE International Conference on
Conference_Location :
Tokyo
Print_ISBN :
0-7803-5731-0
DOI :
10.1109/ICSMC.1999.815606