Research On Modeling Building Of Digital Library Personalized Recommendation Based On Ontology

The digital library as the important information infrastructure and a new knowledge source brought convenient for the users as well as trouble for them too.When users faced these mass information only to accept them passively,and then submerged by huge amounts of information. For information providers,they just provided unified resources and services,could not satisfy all users’ personalized needs. It might led to the“information overload”and“knowledge hunger”,the ineffcient allocation of information resource distribution,unreasonable information flow. Thus,the digital library not only need response to users’ request timely and accurately,but also met different users’ personalized needs forwardly. Then the research of digital library personalized recommendation caused great concern in home and abroad.The personalized recommendation of digital library was based on user needs,analyzing users’ behavior,understanding their preference and providing personalized information service with high efficiency. The personalized recommendation of digital library could establish a good interactive platform. Thus recommender system could timely find users interest resources according to the change of user preferences,then to win their confidence. But the traditional personalized recommendation of digital library was based on keywords and transaction databases.so it had some problems as follow:the inadequate accuracy of the users’ interests mode precision;the description of resource feature hard to match with user interests,lacking flexible personalized service;the characteristics of information resources description mechanism lacking of semantic support,and hard to match the users’ interests model precisly;the low accuracy of computing of Sparse matrix;difficult to find new joining projects and so on.Based on this,this paper introduces the ontology technology into digital library personalized recommendation to realize building the semantic users’ interests model,describing the digital library information resource, designing the personalized recommendation algorithm based on ontology and constructing the digital library personalized recommendation model based on ontology.Finally it realizes the accurate description of user’s interests based on ontology. This paper use protégé to construct digital library domain ontolog based on the existing glossary. And then it analyzes the digital library domain ontology which was described by OWL,extracts the concept of users’ interests to build its ontology,dynamic traces changes of users’ interests and reasonable evolves users’ interests model. Based on digital library information resources(mainly journals and books)describe the semantic of digital library information resource. Based on the digital library information resources which is recommended by the collaborative filtering recommendation algorithm provide knowledge service.The new ideas bring forth in the paper lies in:Firstly,it designs the digital library personalized recommendation model based on ontology,Aiming at problems of digital library personalized recommendation to analyze the general recommendation model of digital library and introducing the the ontology technology to design the overall architecture of the digital library personalized recommendation and some related functional modules;Secondly,it introduces ontology technology to optimize the traditional personal recommendation algorithm.For this paper,there are six parts:Character I introduces the purpose and meaningness of this topic,the research status and innovations of this paper.Character II describes related concepts and technology’s characteristics,and analyzing general recommendation model,traditional recommendation technology and the types and characteristics of digital library resources.Character III analyzes related basic theories of ontology technology(concept,classification,function,building principle,of languages and tools and the process of construction), as well as explains the application of ontology in personalized recommendation technology.Character IV designs a digital library personalized recommendation model based on ontology from the construction of ideas,the overall framework of constructing the model and the process of recommendation.Then,it describes the ontology-based resource libraray of the user information and resources of digital library and the ideas of personalized recommendation algorithm.Character V implements some functions of digital library recommended model. Specially,choose appropriate solutions and build an ontology resources combined with specific tools.Character VI mainly concludes this paper and look forward the next research work.