改进型协同过滤的图书推荐算法

A Book Recommendation Algorithm Based on Improved Collaborative Filtering

针对用户从海量图书中选择喜欢图书较难的问题,提出一种基于图书属性分组的改进协同过滤算法。该算法首先根据用户喜欢的图书类型去选择相似用户,缩小数据集,再根据基于用户的协同过滤算法寻找最近邻居集合,然后根据项目推荐值的方法向用户推荐感兴趣的图书序列。实验结果表明:在同一数据量下,该算法在推荐数据量以及覆盖率方面均优于同类算法。

In order to solve the problem that users are difficult to select their favorite books from a large number of books, a collaborative filtering algorithm based on book attribute grouping is proposed. The method first selects similar users according to the type of books users like, then reduces the data set, and then finds the nearest neighbor set according to the collaborative filtering algorithm based on users. Then according to the project recommended value method to recommend the user interested in the sequence of books. The experimental results show that the proposed algorithm is superior to the same algorithm in the recommended data volume and the accuracy under the same data volume. The algorithm improves the user satisfaction.