基于差异时效的服务评价模型

针对KNN(k-nearest neighbor)方法在服务评价过程中存在的时效量化、评价窗口宽度以及反馈控制等问题,提出了具有差异时效的服务评价模型(WSEM-VTU).在WSEM-VTU中,采用系统动力学方法研究复杂时效量化方法,以获得差异时效量化结果;基于复杂时效量化方法得到的结果,自适应计算评价窗口宽度;根据评价统计特征,设计恶意评价反馈控制策略.通过实验,将WSEM-VTU与现有评价模型WSEM-E及WSEM-KNN进行比较,结果表明:WSEM-VTU的平均误差为0.877,比WSEM-E和WSEM-KNN分别降低了1.020和0.135;引入反馈控制策略后,WSEM-VTU出现恶意评价的情况平均降低67%.

Web Services Evaluation Model Based on Variant Time Utility

To solve the problems of the KNN(k-nearest neighbor) method in the quantification of time utility,the width of evaluation windows and the control of feedback,a web services evaluation model with variant time utility(WSEM-VTU) was proposed.In WSEM-VTU,the system dynamics is used to investigate the complex quantification method of the time utility to achieve the variant time utility.From the results of the quantification method,the width of the evaluation windows is adaptively calculated.In terms of the statistics of feedbacks,a feedback control strategy is involved.By means of experiments,WSEM-VTU was compared with the current evaluation models,WSEM-E and WSEM-KNN.The experimental results show that the average error of WSEM-VTU is 0.877,being lower 1.020 and 0.135 than the errors of WSEM-E and WSEM-KNN respectively.With the feedback control strategy,malicious ratings are reduced by an average of 67%.