Web Text Mining And Information Retrieve Services Based On Ontology

Ontology is an explicit specification of a conceptualization system. In essence, ontology is a set of the concept of a professional field, as well as a collection of the relationship between these concepts. This paper applies the ontology into the processes of text mining. The purpose is aimed to use the semantic description of ontology to represent the characteristics of text content, and then the text mining based on the semantic level will go on. The experiment proved that this method can improve content retrieval precision rate and recall rate, and enhance the results that can be explanatory. Ontology-based text mining has become an important research topic.This paper has designed ontology of missile. It could accumulate 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 web text mining and information retrieve service prototype system, the text mining effectiveness is improved, the text retrieve based on semantic is also implemented and the solid foundation for the further research on ontology-based text mining is built.The innovation of this article:(1) Developed a Chinese Ontology Manager based on OWL framework. This Ontology Manager not only has the features of modifying,updating the ontology, but also can analyze the OWL file to get the ontology knowledge. The ontology knowledge could be shared and reused. Improved the classical Vector Space Model based on the semantic context from ontology. The use of the concept of vector space model to represent the text make the text mining have more semantic context, and the accuracy of clustering was improved.(2) After text mining finished, the mutual information statistical algorithms was used to identify the new ontology terms associated with the ontology concepts to achieve the expansion of the ontology.