Research Of Learning Resource Recommender System Based On Ubiquitous Learning

With the swift development of Internet and computer technology,digital education plays an increasingly important role in people's learning and life.In the current information environment,the contradiction between tremendous educational resources and users' rapid access to personalized learning information resources is becoming increasingly obvious,which leads to ubiquitous learning evolving into ubiquitous search.Introducing personalization recommendation algorithm into education domain,and combining it with pervasive learning materials can help users receive individualized ubiquitous learning resources.Therefore,this paper makes a comparative analysis of the popular personalized recommendation algorithms,and combines the features of pervasive learning resources,uses personalization recommendation algorithm to recommend learning resources for learners.The main work of this paper is as follows:(1)With the explosive growth of the number of network resources,the information obtained by the traditional search engine only has many recommendation results without noticed personalization,which makes it difficult for learners to obtain learning resources that meet their individual needs.This paper introduces personalized recommendation technology into the research area of education combined with ubiquitous learning resources,which can help learners to obtain personalized ubiquitous learning resources.(2)This paper also proposes an effective migration learning algorithm C-CMF,which aims to learn knowledge from the auxiliary binary evaluations with “like” and “dislike”information and migrate them to the target task of numerical scoring by using the auxiliary data user characteristics and the interaction with the information of the target user characteristics,we can obtain efficient learning ability,and improve the performance of target numerical prediction,and largely solve the cold start problem in the traditional collaborative filtering algorithm.(3)Trough a wide range of experiments,we verified the effectiveness of the transfer learning algorithm C-CMF that proposed in this paper.Based on that,we design and implement a prototype of personalized learning resource recommendation system based on ubiquitous learning.