The existing Web service QoS prediction methods are always based on other consumers' experience, which assume that the QoS data coming from these consumers are effective and trustworthy. If the QoS data coming from these consumers are not reliable, the results of QoS prediction method are not precise. A Web service QoS prediction method based on reputation was proposed. First the basic conceive and main process of this method were introduced, and then the classification of service consumers' reputation, the calculating method of service consumers' reputation and the prediction method of web service QoS were stressly analyzed. Finally, a prototype system was implemented to evaluate the proposed method and the traditional arithmetic average method through simulation experiments. The results show that with the proposed approach the preciseness of QoS prediction for Web services can be improved significantly.%基于现有的服务质量预测方法大多是以其他用户的使用感受为依据,但前提部是假设这些信息是准确可信的,一旦这个假设难以得到保证,那预测的结果将会出现重大偏差.从服务用户信誉度的角度,提出一种Web服务质量预测的方法.首先介绍该方法的基本思想与主要过程,随后对评价用户信誉等级划分、评价用户信誉度计算以及服务质量的预测算法等内容进行重点分析:最后,通过仿真实验,将本文提出的方法与传统的算术平均值法进行比较.研究结果表明:该方法能有效提高Web服务质量预测的准确性.
展开▼