基于知识图谱的教育信息资源个性化推荐方法  

Personalized recommendation method of educational information resourcesbased on knowledge graph

作　　者：覃湘荔 Qin Xiangli(Guangxi Modern Vocational and Technical College,Hechi 547000,China)

机构地区：[1]广西现代职业技术学院,广西河池547000

出　　处：《无线互联科技》2024年第8期80-82,共3页Wireless Internet Technology

基　　金：2023年度广西现代职业技术学院校级立项科研项目,项目名称:壮族民间游戏融入幼儿园教育活动的实践研究,项目编号:GXXDZD202304。

摘　　要：现有推荐方法的推荐准确程度低,文章提出一种基于知识图谱的教育信息资源个性化推荐方法。首先,通过协同过滤采集教育信息资源使用者的行为数据,获取用户的兴趣偏好。其次,运用知识图谱计算使用者兴趣和资源实体属性匹配度,获得资源实体与语义关系属性匹配度。最后,通过匹配使用者和教育信息资源特征,比较不同推荐结果隶属度值,判断与使用者兴趣的匹配程度,产生个性化推荐结果。实验结果表明,实验组的稀疏度为80%,推荐准确率为96.3%,所提方法达到良好的匹配效果,可以提供更加精准的推荐内容。Due to the low accuracy of existing recommendation methods,a personalized recommendation method of educational information resources based on knowledge graph is proposed in this paper.Firstly,user behavior datas of educational information resources are collected through collaborative filtering to obtain users’interests and preferences.Then,the knowledge graph is used to calculate the matching degree between user interests and resource entity attributes,and obtain the matching degree between resource entities and semantic relationship attributes.Finally,by matching the characteristics of users and educational information resources,comparing the membership values of different recommendation results,determing the degree of matching with user interests,the personalized recommendation results are generated.The experimental results show that the sparsity of the experimental group is 80%,and the recommendation accuracy is 96.3%,the proposed method can achieve the good matching effect and provide more accurate recommendation content.

关 键 词：知识图谱 教育信息资源 个性化推荐 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]