The Study And Implementation Of Vertical Search Engine Oriented On The Car Subject

With the rapid development of network technology, knowledge of the internet expansive growth, more and more internet users prefer to use search engines for information retrieval. Internet users use the most general search engine that is Baidu, Google, etc. This kind of search engine is covers a wide range of knowledge and returns a lot of associated knowledge, but it doesn’t meet the needs of users when search a field of knowledge. The main reason for the phenomenon is the amount of information and the query inaccurate. Vertical search engines emerge as the times require.However, the vertical search engines still use the keyword-based information retrieval way, which is lack of semantic association and unable to realize the semantic retrieval. The reasons for the problem that the collection documents of search engines are lack of semantic annotation. Therefore, search engine cannot do semantic analysis about user’s query. The appearation of ontology can solves information annotation, realize semantic retrieval. Combine ontology with search engine to achieve semantic retrieval function.This paper studies the basic conception, the vertical search engine features and working principle. At the same time, this paper deals with the present situation of search engine at home and abroad, the application of ontology technology and knowledge in vertical search engine. The main works of this paper as follows:(1) Analysis of system structure of Heritrix, expand the link processor of Heritrix to achieve the purpose of customized crawl; According to the problem of Heritrix download speed slowly, introducing hash algorithm aim to complete multi threads and high speed download; in the information extraction, using the HTMLParser to construction extractor, aim to extract specific website and obtain the structural information.(2) Making a contrast on the methods of building ontology, using the Protege3.2.1and the OWL of ontology description language to complete the car products domain ontology. Using ontology concepts, ontology instances, ontology properties to achieve a semantic query algorithm.(3) In-depth study of the ontology reasoning machine and its realization technology, realized the reasoning function for car ontology with more popular Jena reasoning machine by comparing four kinds of reasoning machine.Finally, develop a vertical search engine based on car ontology, the vertical search engine implements ontology-based semantic expansion algorithm. Experiment shows that the vertical search engine on overcoming the problem in keywords retrieval, at the same time, application of ontology in the vertical search engine that can solve practical problems.