基于混合聚类优化协同过滤的Web服务推荐

Web service recommendation based on hybrid clustering optimization and collaborative filtering

为进一步提高QoS预测准确率,提出一种基于混合聚类优化协同过滤的Web服务推荐方法。采用一种新的混合聚类算法替代传统的top-k算法,结合三个阈值来确定目标用户及服务的相似邻居集合;将基于用户和基于项的皮尔逊相关系数预测方法进行加权结合,从而预测目标用户及服务的QoS值。在香港中文大学发布的真实Web服务QoS数据集WS-Dream上进行了实验,实验结果表明,稀疏度为30%时,所提方法的平均绝对误差和均方根误差均低于现有协同过滤推荐方法,显著提高了QoS值预测精度。

In order to further improve the accuracy of QoS prediction,a Web service recommendation method based on hybrid clustering optimization and collaborative filtering is proposed.A new hybrid clustering algorithm is used to replace the traditional top-k algorithm,and three thresholds are combined to determine the similar neighbor sets of target users and services;The Pearson Correlation Coefficient prediction methods based on users and terms are weighted and combined to predict the QoS values of target users and services.The experiment is carried out on the real Web service QoS data set WS-Dream published by Chinese University of Hong Kong.Experiment results show that when the sparsity is 30%,the average absolute error and root mean square error of the proposed method are lower than the existing collaborative filtering recommendation methods,and significantly improve the prediction accuracy of QoS value.