Research And Implementation Of Group Recommendation System For Learning Resources Based On Membership Relationship

Online learning system breaks the limitation of time and space,makes up for the deficiency of traditional learning methods,and greatly promotes learners’ learning.However,due to the surge in the number of educational resources,it is difficult for learners to obtain the resources they are interested in in a short time.In order to solve this problem,researchers apply personalized recommendation technology to online learning system,which greatly improves the efficiency of online learning.In fact,many learning activities of the participants are not individual learner,but in the form of a learner groups to embark,in this case,in order to improve the learning efficiency of learners’ group,meet the needs of recommended to learners’ group learning resources,to build an intelligent learning resources group recommendation system has an important value and significance.In view of the above-mentioned actual needs,in order to better recommend learning resources for learner groups,this thesis summarizes the shortcomings by analyzing the current status of group recommendation algorithms at home and abroad.In reality,when the learner group conducts learning activities,there will definitely be mutual influence among the learner members.This influence can change the learner’s original preference to a certain extent,so the mutual influence among the learner members cannot be ignored.Based on this,this thesis designed and implemented a group recommendation system of learning resources based on member relationship.The system mainly includes three functional modules,namely learner module,teacher module and manager module.The recommendation function in the learner module is the core function of the system.When recommending for the learner group,the mutual influence among the learner members is fully considered,and a group recommendation algorithm based on the relationship between the learner members is proposed.Will first learners into different group of learners,according to the group to calculate the relationship between learners learners’ influence,according to determine the influence learners leader and the leader of a group,and given that the influence of the different weights,and combining with learners to members of the personality,update the learner members affected by the other members within the group after the preference,Then,the mixed preference integration strategy is used to transform the learners’ personal preference into the learners’ group preference,and finally,the resources are recommended to the whole learner group.In addition,relevant experiments are designed to verify the recommendation effect of the algorithm and improve the learning efficiency and satisfaction of learners.The system uses B/S architecture,using Vue and Springboot framework respectively to achieve the front and back end of the system,Mybatis acts as the persistence layer of the system database,the overall structure is clear and reasonable.According to the requirements and detailed design of each module,the function of each module is realized,and the system is tested and improved to ensure that the system can achieve the expected design objectives,so as to meet the needs of learners.