Chang Min Kim, Yong Gi Kim

The KIPS Transactions:PartB , Vol. 8, No. 2, pp. 115-122, Apr. 2001
10.3745/KIPSTB.2001.8.2.115,   PDF Download:

Abstract

BK-FIRM (Bandler and Kohout's fuzzy information retrieval model) suggested by Bandler and Kohout, uses the concept of fuzzy relation, and it is able to retrieve documents in the way based on not morphology but semantics, dissimilar to traditional information retrieval theories. It also has merits such as building thesaurus automatically and providing fuzzy ranking system. BK-FIRM, however, is not able to apply so many domains because its operations have very high time complexity. In the paper, we introduce A-FIRM (Advanced Bandler-Kohout fuzzy information retrieval model) which improves time complexity of BK-FIRM by using reduced term set, and present A-FIRS (Advanced Bandler-Kohout fuzzy information retrieval system) which is the information retrieval system based on A-FIRM. A-FIRS consists of two parts, preprocess unit and real-time process unit, The first builds the document base and the thesaurus, the last analyzes user requests and retrieves documents. Each process unit consists of 4 procedures. A-FIRS is designed to work with WWW environment, and so it offers retrieved documents, when a user give a query through WWW.

Statistics
Cite this article