Research And Implementation Of Online Service Evaluation Method For Incomplete Ordinal Preference

With the rapid development of Internet technology,online services have been widely used in many fields such as Web services,e-commerce,and search engines.Correspondingly,there are many online services with the same or similar functions on the Internet.Users choose services that meet their needs,which requires a lot of time and effort.Therefore,users usually need to use the third-party perspective-based online service evaluation method to effectively sort candidate services to assist them in service selection.However,because different users have different evaluation criteria,different users' ratings or text reviews for the same service are not comparable,so that such online service evaluation methods based on numerical ratings or text reviews calculating service evaluation results are unreliable and misleading.At the same time,due to the considerable candidate services and users may be unable or unwilling to provide complete preference information about all candidate services due to issues such as cognition and privacy,incomplete preferences have become extremely common and cannot be ignored phenomenon and problem.For this reason,this paper uses ordinal preference as the basic evaluation data to solve the problem that service evaluation results may be misleading and unreliable due to different user having different evaluation criteria.Aiming the issue of online service evaluation problem in the case of incomplete ordinal preferences,this paper proposes an online service evaluation method for incomplete ordinal preferences: First,by analyzing the original definition of Kendall tau distance,a Kendall tau winner distance for complete ordinal preferences is proposed;Then by analyzing the structure of ordinal preferences and calculation characteristics of the Kendall tau winner distance extend the Kendall tau winner distance to the scenario of incomplete ordinal preference,and use it as a service evaluation index to measure the quality of each candidate service to calculate the service evaluation result;Finally,an algorithm and a preference elicitation strategy for evaluating and optimizing the quality of service evaluation results are designed.It is used to quickly improve the quality of evaluation results by applying a preference extraction strategy to the user when the quality of evaluation results cannot meet the requirements.By defining effective service evaluation indicators,as well as quality evaluation algorithms and preference extraction strategies for evaluating and optimizing service evaluation results,the problem of online service evaluation in the case of incomplete ordinal preferences is solved.At the same time,it also ensures that the quality of the evaluation results does not decrease due to the sparseness of the known preferences.The experiment proves the rationality and effectiveness of the method.The experimental results show that the method in this paper can efficiently and stably calculate high-quality service evaluation results,and while ensuring high-quality evaluation results,it can also reduce the amount of user preference data required to calculate the final service evaluation results as much as possible.Finally,based on the online service evaluation method proposed in this paper,a prototype system for online service evaluation oriented to incomplete ordinal preference is designed and implemented.