论文部分内容阅读
[目的/意义]针对移动在线学习平台中用户评价具有布尔变量属性的学习资源,提出一种适用于该类资源的协同推荐方法。[方法/过程]首先采用基于用户自身属性和已有好友分布特征的FRUTAI算法,确定目标用户的最近邻集;然后在解决数据稀疏的基础上,提出适用于布尔型移动在线学习资源的协同推荐方法;最后选取实证对象,依据相关评估方法评估推荐结果。[结果 /结论]在以豆瓣书评网数据作为数据集的实证中取得了较好的推荐效果。实证结果表明,本文提出的改进的协同推荐算法可以有效地应用于移动在线学习平台中的布尔型学习资源,具有较好的推荐效果。
[Purpose / Significance] To evaluate the learning resources with Boolean variables in mobile online learning platform, a collaborative recommendation method suitable for such resources is proposed. [Method / Procedure] Firstly, the FRUTAI algorithm based on the user’s own attributes and existing friends’ distribution features is used to determine the nearest neighbor of the target user. Then, based on the sparsity of data, a collaborative recommendation for Boolean mobile online learning resources Method; Finally, select the empirical object, based on the relevant assessment method to evaluate the recommended results. [Results / Conclusions] Good results have been obtained in the empirical study using the data from the book review database. The empirical results show that the improved collaborative recommendation algorithm proposed in this paper can be effectively applied to Boolean learning resources in mobile online learning platform, and has good recommendation effect.