Research On Ontology-Based Web Intelligent Retrieval

With the rapid development of Internet, Web has developed into a global information resource library. Information retrieval system must solve the issue how to retrieve required information from massive information source precisely and quickly. Most of the existing search engines are either full-text search based on keywords or category browsing based on site themes and are lack of semantic processing ability. That often leads to inefficient search result. The reason for the above problem is that user demands structural and knowledgeable results while the Web resource is disordered and unstructured. Faced with the above-mentioned problems, improving the quality of Web information retrieval and searching for new and intelligent retrieval method have become important topics in information retrieval and data mining fields.Ontology is a formal specification of a shared conceptualization. In essence, it is a collection of the concepts in the field as well as the relationship between them. Using it to expand and reason the query expression, the accuracy of retrieval process could be improved, thus the retrieval process could be enhanced to the semantic level. Therefore, ontology-based information retrieval has become an important research topic.This paper introduces the principle of ontology and research on it both at home and abroad. An experimental ontology of the military aircraft field is built by utilizing Protégé. In order to solve the problem of traditional information retrieval based on keywords, a domain ontology-based semantic information retrieval model is proposed. On the basis of establishing the relationship between the ontology terminology and the content of document, firstly the user's query expression is pretreated, then the similarity is calculated using domain ontology, and finally the sorted relative documents corresponding to the query are shown. The built ontology-based Web intelligent retrieval prototype system validates the effectiveness of the model preliminarily.This paper accumulates the experience and method for constructing ontology of the national defense field, and it has some exploration significance and research value for forming methodology of constructing mature ontology in national defense field. By realizing intelligent retrieval prototype system, the retrieval effectiveness is improved and the solid foundation for the further research on ontology-based intelligent retrieval is built.