Research On Resources Personalized Recommendation Strategy Based On Ontology In E-learning Environment

In recent years,with the rapid development of E-learning,the amount of educational resources in the Internet also increased in the form of an index.The massive growth and non-standardized expansion of information has brought about the problems of "information overload","information island" and "digital ruin".People enjoy the Internet to bring rich resources at the same time,but also had to face vast amounts of data and a lot of spam.From the education itself,the various colleges and universities education information systems are fragmented,there is no uniform technical standards,neither plan nor understand the real needs of the market,which led to the use of online education resources is very low,educational resources did not play its proper the value of.LRMI metadata as a model of value-based learning resource description program provides a potential solution for future learning analysis and the development of open educational resources.The resource of the personalized recommendation system through the independent filtering of massive resources,screening out the learners need learning resources,reducing the cost of learning resources for learners,is an effective way to solve the problem of information overload.Personalized recommendation system is the most successful recommendation technology is collaborative filtering recommended technology,but it also exists sparse data problems and cold start problems,etc.,recommend the recommended quality of the system to be improved.In view of the above problems,this study puts forward the personalized recommendation model of resource based on ontology in E-learning environment combined with the development trend of current educational resources and resource individuality recommendation.By analyzing the existing metadata specifications at home and abroad and the personalized recommendation algorithm of domestic and foreign resources,this paper chooses the appropriate metadata extraction technology and Chinese semantic thesaurus,and designs the personalized recommendation model of E-learning platform The Finally,based on the undergraduate public curriculum examination of English as an example,the design of the platform for the design of personalized examples of resources.The main contents and research results of this paper are summarized as follows:In the background of modern distance education,this paper studies the relevant theoretical basis and technical foundation of the individual recommendation of learning resources,and summarizes the research status of personalized recommendation of learning resources.Ontology,semantic network,metadata and resource individuality recommendation theory are gradually introduced from other fields into the field of education,providing a very convenient way for the management and utilization of learning resources.2,this paper studies the value-based learning resource description specification-RRMI metadata,summarizes the current trend of open educational resources and the reality of resource search,and puts forward the use of LRMI metadata specification for domestic E-learning learning platform metadata Comprehensive development of educational resources to prepare.3,the design of the domestic universal E-learning environment based on the ontology of learning resources personalized recommendation model.By analyzing and summarizing the advantages and limitations of different kinds of resource recommendation algorithms,and combining with the actual demand,we design the learning resource recommendation algorithm suitable for E-learning platform.Finally,the calculation steps of resource recommendation are shown in the flow chart.4,based on the undergraduate public curriculum as an example to design English based on the ontology of learning resources personalized recommendation program.This paper analyzes the development process of the recommendation system from the four aspects of data acquisition,data distributed storage,data classification and aggregation calculation,and data resource recommendation.