The Learning Resource Recommendation For The Learning Community Based On Learning Analytics Technology

People's learning concept has been changing greatly with the advancement of educational information and the establishment of virtual learning communities.Many people take participation in learning activities by plenty of learners in the form of groups,such as BBS forum,QQ group,and Micro groups and so on.People gradually formed an idea of learning that learning was participating in social group by themselves and completing the knowledge construction process of learning,and therefore learning by "online learning community" in the form of online to be a new trend.Now,The community member could rarely figure out his own interesting resources,even study effectively,since the continuous growth of learning resources and the overload of information which could be cognitive maze and distraction,just like,it is hard for student to choose right resource because there are a lot of resource without valuable information.Recommended technology is considered to be one of the effective ways to solve the problem of information overload and cognitive maze.However,the recommended methods of personalized learning resources are mostly for individual learners.Individualized learning resources could not meet the common needs of learning community members for learning activities.In addition,it was not suitable for traditional recommendation technology to give the recommendation resources since the learning score data was very scarce in online learning platform,that is to say,making the traditional way of resource recommendation was blocked.Learning analysis which is an analytical tool for learning data and compile the recommended technique to calculate the predicted score from multimensional to improve the recommended effect could extract valuable data from learning behavior.Therefore,this study intends to recommend resources from the perspective of learning analysis,analyze the data of the members in the Learning community by using the learning analytic technology.The above data is about the learning behavior that we could collect it from the learning platform.The members' learning preference is obtained basing on the BP neural network model.On this basis,use The Factorization Machine to obtain the list of recommended resource,use the weighted model to get the group recommended list finally,and Implement the TOP-N recommendation.This study mainly adopts the methods of literature research,data mining and off-line experiment verification.This paper analyzed the relevant research of learning analysis and group recommendation,and finds out that learning analysis technology which has a good development trend has developed with the innovation of science and technology and the innovation of education through the literature research methods.This paper studied the Factorization Machine,the key technology of group recommendation,verified the application value of BP neural network model and the Factorization Machine in this research.Finally,the recommendation results of the group recommendation model were verified by offline experiments.It was found that the group recommendation based on the learning analysis technique could meet the learning needs of the learning community and improve the learning efficiency of the members.