Research And Implementation Of Online Service Evaluation Based On Plackett-Luce Model

In recent years,with the rapid development of Internet technology,the scale of online services has grown significantly.There are a large number of online services with the same or similar functions in the Internet.Users need to spend a lot of time to choose the online services they need in many services.The online service evaluation method can help users to better choose online services.The existing research methods all assume that users have the same evaluation criteria,but because the user's consumption hobbies and consumption habits are different,each user's evaluation criteria for online services are different in practice.The user ratings of the same service obtained under different evaluation criteria are not comparable.The online service evaluation results obtained according to the incomparable online service scores cannot reflect the real performance of the online services,and misleading users to select online services may occur.In order to solve the above problems,this artical adopts an online service evaluation method based on Plackett-Luce model,which is based on the inconsistency of user evaluation criteria,and converts the incomparable online service score into the final online service evaluation result by probability method..Firstly,the Plackett-Luce model is used to obtain the preference value of each user for different online services based on the user service score.Secondly,under the same user evaluation criterion,each user's preference relationship for all services is obtained according to the preference value;The number of times of each online service is normalized and standardized under all user preferences.Finally,the log likelihood function and the iterative function formula of the Plackett-Luce model are established and the final evaluation of each online service is obtained by the value of the iterative dominant times.result.Ranking online services based on the results of the evaluation and helping users select online services.If there is an optimal ranking,the online service selected at each stage of the Plackett-Luce model online service evaluation method is the online service with the highest ranking probability,so the online service evaluation result is the closest to the real ranking result..The experimental results show that the proposed method can solve the problem of incomparable user service scores when the user evaluation criteria are inconsistent.At the same time,the experimental results show that the proposed method satisfies the inverse symmetry,monotonicity,and anti-manipulation,Kendall tau distance and most of the criteria.Therefore,the online service evaluation results obtained by this method are effective.Finally,based on the online service evaluation method proposed in this artical,a complete online service evaluation system is designed.