融合学习者社交网络的协同过滤学习资源推荐

发布时间：2018-06-21 07:49

  本文选题：社交网络 + 协同过滤　； 参考：《现代教育技术》2016年02期

【摘要】：传统的协同过滤推荐算法存在冷启动和数据稀疏的问题,使得新学习者因历史学习行为记录稀疏或缺失而无法获得较准确的个性化学习资源推荐。鉴于此,文章提出将学习者社交网络信息与传统协同过滤相融合的方法,计算新学习者与好友之间的信任度,借助新学习者好友对学习资源的评分数据,来预测新学习者对学习资源的评分值,以填补新学习者在学习者—学习资源评分矩阵中的缺失,实现对新学习者的个性化学习资源推荐。实证研究结果表明,该方法在一定程度上能够解决传统协同过滤方法的冷启动和数据稀疏问题,提高个性化学习资源推荐的准确率。
[Abstract]:The problems of cold start and sparse data in the traditional collaborative filtering recommendation algorithm make it impossible for new learners to obtain more accurate personalized learning resources recommendation due to sparse or missing history learning behavior records. In view of this, this paper proposes a method that combines the information of learners' social networks with traditional collaborative filtering, calculates the trust between new learners and their friends, and makes use of the new learners' friends' scoring data for learning resources. To predict the new learners' scores on learning resources, to fill the gaps in the Learner-Learner Resource scoring Matrix, and to realize the personalized learning resources recommendation for the new learners. The empirical results show that this method can solve the cold start and data sparse problems of traditional collaborative filtering methods to some extent and improve the accuracy of personalized learning resources recommendation.
【作者单位】： 湖北大学教育学院;
【基金】：教育部人文社会科学研究青年基金项目“基于互动电视的课堂教学模式与策略研究”(项目编号:14YJC880109)阶段性研究成果
【分类号】：G434
，

本文编号：2047796