关键词: Web服务, 服务质量预测, 协同过滤, 高斯归一化

Abstract: To deal with the prediction error caused by different QoS data range in traditional collaboration prediction approaches,a collaborative Web service QoS prediction approach was proposed based on Gaussian normalization theory.QoS data in different range were mapped into a uniform interval by utilizing user-based and item-based Gaussian normalization formula.Similarity rating was computed by utilizing Pearson correlation coefficient,and a minimum similarity threshold factor was introduced to filter weakly correlated neighbors.Missing values in normalized QoS matrix were calculated by collaborative filtering approach.Furthermore,a restoration algorithm was designed to restore the corresponding value in origin QoS matrix,a strategy that systematically combines user-based approach and item-based approach was adopted to make comprehensive prediction.Contrast experiment results in real datasets and case verification showed that the proposed approach could effectively improve the prediction accuracy.

Key words: Web service, quality-of-services prediction, collaborative filtering, Gaussian normalization