The Research On Ontology Evaluation Based On Search Ranking Algorithm

Ontology, as the core technology of the Semantic Web, is an advanced knowledge representation in the domain of artificial intelligence. It reflects the nature of things and phenomena by clearly defining the concepts and the relationship between them, and plays an important role in the knowledge acquisition, representation, analysis. Many research institutions at home and abroad have carried out extensive research.With development of technology in construction and evolution of the ontology, the application of ontology is widely used. As the large number of ontology, all kinds of domain ontology emerging in the same domain, differences of applicability in all kinds of domain ontology, we urgently need an ontology evaluation system help user to make a chioce. Currently, research in ontology evaluation is still in the initial stage, and the evaluation system, evaluation indicators, evaluation methods of the ontology need to be further improved.This paper presents an ontology evaluation system based on model dynamic search ranking algorithm. This model achieves the online evaluation of ontology and provides three major functions which are online uploading, online searching and online evaluation of ontology. We provide users with an evaluation way which combines dynamic and static also combines subjective and objective. The main work of this paper is as follows:1) We analysis the relative theory of ontology, current situation of ontology research, importance, necessity and urgency of domain ontology evaluation system. We summarize achievement in the current ontology evolution research areas, minds and methods of ontology evaluation experts, and existing problems in current ontology evaluation system.2) We present a dynamic search ranking algorithm, in the dynamic process of ontology searches, the system automatically sort the ontology by the degree of user queries. Given a set of discovery rules of ontology document, we improve the efficiency of data collection in ontology search, improve the traditional ranking algorithm of ontology, not only in accordance with the concept and importance of the relationship sort of ontology, but also from the perspective of ontology intelligent search user search query order to calculate popularity. 3) We propose a static ontology evaluation method, a series of evaluation indicators of the ontology is given, methods and algorithms of translatability, accuracy, comprehend-siveness, coverage, cohesion, coupling in ontology are also given.4) We design ontology evaluation system combining objective evaluation and subjective evaluation. Through an open user voting system, the user can subjectively evaluate the ontology. Ontology users are divided into three categories:ontology engineers, domain experts, application system users, and they subjectively evaluate the rationality of ontology construction, the correctness of ontology concepts, and the applicability of application system.