The Research Of Expanding The Semantic Information Function To Search Engine

With the rapid development of Internet, Web information is rapidly increasing, people are more dependent on the Internet through the search engine to obtain information on. Unfortunately, most search engines are used in a simple matching of keywords to search, and without deep understanding of semantics into the surface of the user's true intentions. Therefore, how to better understand the user's search request, provide the semantic associative search method has become the new generation of search engine one of the problems solved.For the semantic search engine key technologies in a more in-depth research, this article mainly include the following:1.Proposed a semi-automatic ontology based on WordNet dictionary construction method. At present, most of the ontology construction methods are constructed by artificial means, artificial construct ontology is a difficult process, not only requires a lot of manpower, material,and the whole process requires the participation of experts in the field. Since the concept of the field of relationships between concepts and definitions is a tricky thing, so to completely automate the generation of ontology rather difficult. This article proposes using existing WordNet a semi-automatic ontology generation method.2.Proposed ontology of the definition of the concept of which is passed between transitivity and inheritance relations are described. Ontology consists of five basic modeling primitives, namely, the concept, relations, functions, axioms, instances of modeling primitives through the analysis of these five are given some formal definition of ontology, and described the transfer and inheritance relations sex, according to the logic of these relations properties, can be inferred that the concept of search required.3.Proposed mechanism for logical inference relationships, the concept of ontology is given is_a and part_of relations between the inference rules.These two inference rules, we can build a good and the concept ontology library to find out there is_a and part_of relations concepts. 4.Proposed an added part_of relational inference based on the ontology search engine model,it can carry on semantic understanding of Web pages from the semantic level and the semantic extension and analysis for the user inquiry condition, so that more of the user's needs are met.