Research On Computing Concept Relatedness Sasead On Ontology

In the information age, Chinese information processing has become an important area of research. The relationship between any two objects is very complicated, we call that they are relational, relatedness is a quantitative representation of relevance. Concept relatedness is widely used in information retrieval, semantic extension, text classification, machine translation and other fields. It is an important technology in the field of natural language processing. So measuring the concept relatedness efficiently is an important task.Currently, there are many researches on calculating the relatedness between words, such as the method based on co-occurrence probability, or the similarity of semantic dictionary. On the one hand, These methods don’t fully considered semantics of the words, on the other hand they don’t analyze the relationship between concepts. For the above problems, the paper do the following works.(1) The research background and classical research methods of computing concepts relatedness are introduced briefly in the paper. Besides We mining the factors that influence the value of relatedness according to the classical calculation method.(2) Study the process of building domain ontology and build An experimental fragment of Confucian-oriented areas is built based on the Seven-step method. Besides, how to map the ontology to a relational database is carried out.(3) Concept relatedness consists of similarity and relevance, which should be computed differently. Introducing the basic relationship between concepts in the domain ontology, We proposes and implements a formula of computing the similarity between ontology which based on semantic distance, hierarchy depth, public ancestors set and the basic relationship.(4) A method to measure the specific association relation besides basic relations is Given. Incorporating both similarity and relevance, we get an overall formula of computing concept relatedness. (5) The research of computing concept relatedness is applied to the semantic extension. For problems of query and retrieval existing in the database, By extending the semantics of key words, retrieval performance of the database is improved.Concept relatedness is widely used in many fields, it can show the degree of relationship between two concepts. So it is suitable for the field of information retrieval. By extending the semantic of the keywords, Search engines can query target clearly, thus improve the recall and precision of retrieval.