Research On Vertical Retrieval Engine Based On Ontology

With the development of Internet, the quantity of Web information resources becomes larger and larger. Confronted with so much information, How to ensure that the information in full, timely and accurately becomes the key issue of the field of search engine. Although general search engines on quantity aspect have more advantages, yet traditional information retrieval technology is mainly based on keyword matching and has little semantic inferring ability. Moreover it does not provide semantic guidance for users, so general search engines are poor on quality aspect. So information retrieval System may miss some information that users really want and return some information that users don't want. The current user's query is being in the direction of " specialized, precise, deep ". How to improve the quality and efficiency of information retrieval becomes an important study field of information retrieval (IR).Vertical search is to provide a specific domain with the certain value information and related services, and the ontology is the abstract and description of concepts, relationships, attributes of the knowledge. Therefore, the combination of ontology and the search engine is an important mean to research vertical search. So how to construct the area of the body, and how to integrate ontology theory with vertical search technology have become the focus of the study.In the beginning, this thesis talks about the basic concepts and principles of the search engine. It focuses on the key technologies of vertical search engine, including Focused-Spider, Structured Information Extraction and Information Retrieval. Then it introduces some theories about the ontology. It expounds the significance and necessity of combining the ontology with the vertical search engine, and focuses on feasibility study combined with examples.This paper mainly focuses on the following work. It develops the domain ontology of computer accessories with Protege and OWL. It focuses vertical search engine based on ontology, and then proposes a kind of framework of ontology-based spider integrated with Pages Similarity Algorithm. To use the ontology of the domain ontology to the page text documents after pretreatment for the semantic annotation based on ontology, and do the structured information extraction of the contents of the documents. To do the semantic extension for the query based on the domain ontology of computer accessories. It focuses on the solution for the query of the ordinary word which is not in the ontology.Finally, we develop a vertical search engine system based on our constructed ontology, and we describe the implement of the application of ontology to vertical search, we find that the semantic retrieval system based on ontology is more comprehensive and exact than the retrieval method based on keyword in the traditional way by analyzing circulation of the system. So it is very useful to research the issue of ontology-based vertical search engine.