一种基于图卷积神经网络的在线课程推荐系统

An online course recommendation system based on graph convolutional neural network

由于在线课程学习不受时间和地点限制,越来越受到广大求学者的青睐,但各大在线教育平台推出的在线课程数量较多,使得用户难以选择。课程推荐是解决“信息过载”的重要手段,然而现有的课程推荐模型对用户和课程隐式交互数据挖掘不足,为此,文中提出一种基于图卷积神经网络的在线课程推荐系统。首先利用用户和课程的多种交互行为分类构建用户-课程二部图;然后将课程知识信息融入用户-课程二部图,利用图卷积神经网络高阶连通性递归地在图上传播嵌入信息,深入挖掘“用户-课程-知识”的关联关系,并设计高效的在线课程推荐系统,迅速响应用户课程请求;最后选取三种经典的神经网络推荐模型进行对比分析。实验结果表明,所提方法具有较高的推荐准确率。

Online course learning is increasingly favored by scholars due to the fact that it is not limited by time and location.However,major online education platforms offer a large number of online courses,making it difficult for users to choose from.Course recommendation is an important means to solve the problem of"information overload".However,the existing course recommendation models are insufficient for data mining of implicit interaction between users and courses.Thereforce,an online course recommendation system based on graph convolutional neural network is proposed.A user-course bipartite graph is constructed by classifying multiple interaction behaviors between users and courses,and the course knowledge information is integrated to the user-course bipartite graph.The high-order connectivity of graph neural networks is used to recursively propagate embedded information on the graph to deeply explore the association relationship of"user-course-knowledge"and an efficient online course recommendation system is designed to quickly respond to user course requests.Three classical neural network recommendation models are selected for the comparison experiments,and the experimental results show that the proposed method has higher accuracy of recommendation.