基于支持向量机与事例推理的Web服务QoS动态预测方法研究

Research Method on Web Service QoS Dynamic Prediction Based on Support Vector Machine and Case-based Reasoning

在开放的网络环境下,Web服务的Qo S具有很强的动态性,而如何准确地预测未来一段时间之后Web服务的Qo S,关系到服务选择与组合的成败,是服务计算领域亟待解决的关键问题.针对此问题,在考虑Web服务负载、任务特征与服务Qo S相互关联的情况下,给出一种基于支持向量机与事例推理的Web服务Qo S动态预测方法.本文首先采用支持向量机对Web服务在一段时间之后的负载进行预测,然后,根据以上预测得出的负载结果和所要处理任务的特征信息,采用事例推理方法对Web服务处理某一具体任务时的Qo S进行预测.实验结果表明,该方法是可行的、有效的,并在一定程度上提高了Web服务Qo S的准确性.

The quality of service（ Qo S） of the Web service has a strong dynamism due to the openness of the network environment in which Web service is located. Howto accurately predict the Qo S of Web service after a period of time has a close relationship with the reliability of service selection and composition,and has become a key scientific issue that needs to be solved urgently in the field of services computing. In viewof the above issue,this paper proposes a method to predict the Qo S of Web service by taking a full consideration for the correlation between environmental factors and the Qo S of Web service. This method firstly predicts the load of Web service in a moment of future based on SVM,and then combined with the load and information of tasks need to be processed it applies CBR to predict the Qo S of Web service which is processing some specific tasks. Experimental results showthat this prediction method can improve the accuracy of Web service Qo S greatly compared with existing methods. The results provide a reliable basis for the objective evaluation and successful Web service composition.