Hybrid query processing for personalized information retrieval on the Semantic Web

摘要

This paper suggests a hybrid query processing method for the effective retrieval of personalized information on the Semantic Web. When individual requirements change, the current method of query processing requires additional reasoning for knowledge to support personalization. To minimize this problem, the hybrid query processing method uses both the query rewriting method and the reasoning method. This paper distinguishes knowledge that is frequently changed from knowledge that is not. The query rewriting method is used for frequently changed knowledge; otherwise the reasoning approach is used. The query rewriting method refers to individual requirements to extend user queries instead of conducting inference. To illustrate the advantage of this method, a Personalized Hotel Search System (PerHSS) was implemented, consisting of hotel domain ontology, question-based and answer-based requirements collector, and a personalized hotel search interface using available Semantic Web technologies. This paper reports the results of the performance of a set of query tests and compares the results to those of similar works. The results show that the suggested method is suitable for the efficient retrieval of personal information.