Personalized Knowledge Service Key Technology Based On Linked Data And User Ontology

Today, technological innovation also reflects the status of educational technology in the information system. The networks facilitate the information exchange, cultural dissemination and knowledge sharing, e-learning become the new direction of development of the education. The reason is the knowledge service system can’t effective understanding and tagging the semantic information included in the network resources, the heterogeneous resources can not be organized into management knowledge. The second is not provided persnalized knowledge service for different users, which have different background and learning requirements.Personalized knowledge service use semantic network, data mining, information retrieval, personalized recommendation technology, to research the online user’s learning behavior, to mine the knowledge background, interest, emotion, social relations information. When the user search, question and autonomous learning in internet, through the logical reasoning and semantic extension to clear its learning needs, to retrieval the linked course data, accurately find the related resources, and show to users through the appropriate visualization methods. To provid navigation, recommendation, Q&A, etc., personalized knowledge service for learners.Personalized knowledge service need to establish a efficient, high cohesion and low coupling knowledge service system architecture, in order to achieve the sharing and communication of the field information and user information between service modules. Now, the research included two main stream. service-oriented LMS (Learning Management System) and personal learning environment architecture (PLE, Personal Learning Environment). But no matter which method, each functional module to the form of Web service unit via the Internet to build learning environments, which lead to the services internal links losing, duplicate development, the development of efficiency decreases, but also to make the service lost its integrity, will not be able to achieve a high degree of sharing and interoperability of the field of information and user information between modules. Which will lead to make the personalized and adaptive cannot be effective. Personalized knowledge service also need to set up the user model, which can accurately reflect the personality and behavior characteristics. At present, the factor of the user modeling is not enough accurate, there are state space complexity is high, the prediction of the change inaccurate problems. Therefore, accurate and timely analysis user behavior and social relations, to establish dynamic user model to improve the quality of the individualized service is one of the key. User modeling research based on user behavior characteristics and knowledge ontology, constructing user ontology; Constructing the corresponding logical rules based on the user ontology and knowledge on ontology for semantic reasoning, dynamic perfect the user ontology and clear the user learning requirements.The goal of the personalized service is to provide the right knowledge for each user quickly and accurately, and high efficiency finish the study, to improve the learning efficiency and learning effect. Mass network learning resource semantic organization and personalized knowledge service is a key scientific problems. Using the semantic web technology to solve the network resource sharing and personalized knowledge service is a new exploration with theoretical significance and application prospect, which has important realistic significance and practical value for the semantic web framework of electronic learning system research promotion. Personalized knowledge service based on semantic data and user ontology have the profound social significance for promoting the network learning resources into the semantic level organization and sharing, the knowledge service into the personalized recommendation and semantic search. The research of this paper from the service architecture, the user model, the semantic search engines, personalized recommendation engines, personalized Q&A and personalized custom expansion.(1) Proposed a efficient, high cohesion and low coupling of personalized knowledge service system architecture based on resources and the user’s semantic information;(2) For the factor of the user modeling is not enough accurate, there are state space complexity is high, the prediction of the change inaccurate problems. Therefore, accurate and timely analysis user behavior and social relations, to establish dynamic user model to improve the quality of the individualized service is one of the key. User modeling research based on user behavior characteristics and knowledge ontology, constructing user ontology; Constructing the corresponding logical rules based on the user ontology and knowledge on ontology for semantic reasoning, dynamic perfect the user ontology and clear the user learning requirements.(3) According to the needs of the individual services, deeply research the recommendation of the engine system architecture and the recommendation algorithm. Using the associative retrieval techniques to alleviate the data sparseness problem, and introducing the labels and users of social relationships to improve the user/product similarity calculation method, so as to improve the diversity and accuracy of the recommended results.(3) for different users search needs semantic annotation of document retrieval, case retrieval and associated retrieval adaptive search strategy, and design of the semantic indexing construction method, semantic recognition and extension methods and personalized sorting algorithm to ease the "cognitive overload";(4) According to the needs of search serive, researched on the architecture, flow and retrieval strategy. Puts forward the semantic tagging document retrieval, case retrieval and associated retrieval strategy. Designed the semantic index construction method, semantic identification and extension methods and personalized sort algorithm. According to the problem of the semantic search accuracy, the corresponding semantic extension methods and algorithm, in order to improve the accuracy of the semantic search.(5) For a specific environment of e-learning, proposed the three layer Q&A architecture to gain answer from Q&A library, linked data and learning resources in turn. To build the semantic navigation treebased on domain ontology hierarchy. Through the user’s selected to determine the learning objectives. The selection of the learning content and the generation of learning path through semantic reasoning. Finally, experiments proved the effectiveness of the proposed algorithm. The development of the prototype system and experiments to prove the effectiveness of the proposed architecture. The results of two experiments show that the in-depth study and the proposed method have innovation, feasible and effective.