Research And Application Of Personalized Knowledge Recommendation Technology Based On Ontology

With the rapid development of the knowledge economy, enterprise’s demand for knowledge isgetting more intense, a lot of enterprises have introduced knowledge management system to manageinternal knowledge resources. However, the sharp growth of knowledge resources in system makesusers difficult to find really useful knowledge, and thus fall into knowledge confusion. Personalizedrecommendation technology is one of the most effective ways to solve information overload problem,it can free users from massive resources and help users access to knowledge efficiently. This articleintroduces personalized knowledge recommendation technology into the Railway Wagon DesignKnowledge Management System, with a focus on the research of user modeling method andpersonalized recommendation algorithm. The main research contents of the paper are as follows:(1) Describes the framework of Design Knowledge Management System Based on PersonalizedKnowledge Recommendation Technology. The framework consists of knowledge resources layer,knowledge processing layer, business logic layer and user interface layer. Business logic layer is thecore layer which mainly processes user modeling and personalized knowledge recommendation.(2) Proposes an ontology-based fine-grained user modeling method. The method defines afine-grained user model description on the basis of ontology-based user modeling method, gives userinterest measure method based on browsing behaviors, combines domain ontology to learn user model,and gives dynamic updating method of user model based on user interest attenuation mechanism.(3) Proposes an ontology-based multilayer collaborative filtering recommendation algorithm, andgives the general idea and processing flow. On the basis of ontology-based fine-grained user model,the algorithm combines several user similarity measure methods, gets the nearest neighbors throughmultilayer filtering for neighbor set, and then recommends knowledge based on group interests of thenearest neighbors, and the effectiveness of this algorithm is verified by experiment.(4) Completes the design and development of Design Knowledge Management System Based onPersonalized Knowledge Recommendation Technology. The structure of the main functions is giventhrough analysis of the project background. The paper also describes the system design andimplementation details, and gives part of running interface.