A knowledge-based approach for answering fuzzy queries in XML

XML (Extensible Markup Language) has become the standard for data interchange over the Internet, thus a large number of heterogeneous data can be integrated into the XML document. Users often have vague or imprecise ideas when searching the XML and may not be able to formulate queries that accurately express their query intentions. Therefore, user might like to issue fuzzy queries that consist of fuzzy predicates for possibly retrieving. Based on fuzzy sets theory and the knowledge base related to the application domain, this paper presents an approach which can translate the fuzzy predicates of the fuzzy query into the precise predicates and extends the query criteria ranges in order to provide approximate answers to the user. The fuzzy predicate is first defined by a membership function, and then is translated into the precise predicate by using the α-cut operation of the fuzzy number. The document segments of XML document satisfying the fuzzy query are returned by using twig pattern matching method and finally ranked according to their satisfaction degree. Experimental results identified that the fuzzy query answering method can meet user's needs effectively.