基于云平台的慕课资源协同过滤推荐算法  

Collaborative Filtering and Recommendation Algorithm of MOOC Resources Based on Cloud Platform

作　　者：徐福江[1] XU Fujiang(Faculty Development Center, Taizhou Vocational and Technical College, Taizhou, Zhejiang 318000, China)

机构地区：[1]台州职业技术学院教发中心,浙江台州318000

出　　处：《微型电脑应用》2020年第5期37-39,共3页Microcomputer Applications

基　　金：2017年浙江省教育厅一般科研项目(Y201738149)。

摘　　要：针对当前慕课资源协同过滤推荐算法存在推荐误差大、无法实现在线推荐的难题,为了提高慕课资源协同过滤推荐精度,设计了基于云平台的慕课资源协同过滤推荐算法。首先分析慕课资源协同过滤推荐的原理,提取慕课资源相似度特征,然后引入k-最近邻对慕课资源相似度进行评价,实现慕课资源分类和协同过滤推荐,最后在云平台分布式、并行实现慕课资源协同过滤推荐算法,并与传统算法进行了仿真对比实验。结果表明,相对于传统算法,提出的算法使得慕课资源协同过滤推荐精度得到较高提升,能够解决当前慕课资源协同过滤推荐算法存在的一些缺陷,而且可以实现慕课资源协同过滤在线推荐,实际应用价值也得到了改善。The design of collaborative filtering and recommendation algorithm for MOOC resources is of great significance.In order to improve the accuracy of collaborative filtering and recommendation of MOOC resources,a collaborative filtering and recommendation algorithm based on cloud platform is designed.Firstly,the principle of collaborative filtering and recommendation of MOOC resources is analyzed,and the similarity features of MOOC resources are extracted.Then the k-nearest neighbor is introduced to evaluate the similarity of MOOC resources and realize the classification and collaborative filtering recommendation of MOOC resources.Finally,the collaborative filtering and recommendation algorithm of MOOC resources is implemented in the cloud platform in a distributed and parallel way.The simulation experiment is carried out for the proposed algorithm and traditional algorithms.The results show that,compared with the traditional algorithm,the accuracy of collaborative filtering and recommendation of MOOC resources in this algorithm has been improved greatly.It can solve some defects of current collaborative filtering and recommendation algorithm of MOOC resources,and realize online recommendation of collaborative filtering of MOOC resources.The actual application value has also been improved.

关 键 词：慕课资源 K-最近邻算法 协同过滤 资源分类 推荐精度 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]