Web Services Evaluation Predication Model based on Time Utility

Current predication models fail to consider the time utility of the web services evaluation predication and treat the different historical ratings in the same way. To solve this problem, we put forward web service evaluation predication model based on time utility. In the model, naïve quantification method and complex quantification method are proposed to achieve the distinct and proper time utility for the services evaluation predication procedure. Then, the quantification results are used to optimize the length of the predication windows. Also, feedback control strategy is involved to enhance the robust of the model when facing malicious. Experimental results shows our model would calculate the proper time utility and obtain the lower predication error compared with current predication models. Feedback control strategy is an effective method to reduce the impact of malicious ratings and guarantee the lower predication error compared with the model without the feedback control strategy.