The Research And Application Of Demain Ontology-based Intelligent Searching System

With the rapid development of Internet, World Wide Web has been the main medium for global information. However, traditional search engine usually returns much irrelevant link which the users have to spend much time to rule out. The reason for it is that demands of the structural information contradict with the factual disordered information, and that difference also occurs on the representation of the concepts.Ontology is an advanced technology of knowledge representation in semantic Web technology. Ontology has the capability to represent meaning of the information by a hierarchical structure, and support reasoning. So ontology-based information retrieval is a more and more important topic.The thesis firstly introduced the basic theories of ontology and semantic Web.Then I expounded the search engine's problems it is confronted with and the trend. Aiming at above problems, I deeply researched the MM algorithm and reformed it. The thesis also gave the calculation of domain knowledge based conceptual similarity and conceptual relativity. After that I studied some important facts when using ontology to represent and process semantic information. It contained design and build knowledge base based on demain ontology, the method of searching knowledge base (KB) and make inference in KB. Finally, I designed and basically achieved the domain ontology-based intelligent searching and tracing system. The testing results proved that through the modules of word-cut, reasoning & inquiry and information tracing, the system can bring the useful, interrelated and abundant information. It has much higher performance than traditional method.