基于Weighted-slope One的用户聚类推荐算法研究

Research on User Clustering Recommendation Algorithm Based on Weighted-slope One

针对传统协同过滤推荐算法存在的数据稀疏性以及实时性差的问题,提出一种基于Weighted-slope One的用户聚类推荐算法。该算法首先利用Weighted-slope One算法的思想对初始的用户-评分矩阵进行有效填充,降低数据的稀疏性;然后,结合初始聚类中心优化改进的K-means方法对用户进行聚类,生成相似用户集合,以缩小目标用户搜索最近邻的范围;最后,结合目标用户所属的聚类,利用基于用户的协同过滤算法搜索最近邻居,为目标用户推荐对应的产品。仿真实验结果表明,改进算法可以显著降低数据的稀疏度,同时提升推荐的准确性和实时性。

Aiming at the problems of data sparseness and the poor real- time performance of traditional collaborative filtering recommendation algorithm,a newuser clustering recommendation algorithm based on Weighted- slope One algorithm is proposed. Firstly,the zero items in user- item matrix are filled by Weighted- slope One algorithm. This operation can effectively reduce the data sparseness. Secondly,the users are clustered by the optimized K- means algorithm. This operation can effectively find the nearest neighbor of the target user. Finally,the corresponding products are recommended to the target users according to their nearest neighbors which are found by the users＇ collaborative filtering recommendation algorithm. Experimental results showthat the improved algorithm can significantly reduce the data sparseness,and improve the real time performance and the accuracy of recommendation.