Online Service Reputation Measurement Based on Consensus Clustering
LIU Yan-li1,FU Xiao-dong1,2,YUE Kun3,LIU Li1,FENG Yong1,LIU Li-jun1
1(Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China)2(Yunnan Provincial Key Laboratory of Computer Technology Application,Kunming University of Science and Technology,Kunming 650500,China)3(School of Information Science and Engineering,Yunnan University,Kunming 650091,China)
Abstract：User ratings of services can be regarded as user classification of services,and the service reputation is the result of clustering based on the user classification of services.The consistency relationship between online service reputation measurement results and all user classifications are not considered in the clustering process,which will lead to the lack of rationality of the reputation measurement results obtained by the aggregation classification.To this end,this paper presents a method of online service reputation measurement based on consensus clustering.Firstly,the user classification information of the service is obtained according to the user-service rating matrix.Secondly,considering the possible similarity between the online service classification information,the minimum-cost spanning tree based on the inter-cluster and intra-cluster similarity is established.Then,cutting the minimum-cost spanning tree from the edge with the smallest similarity value and generating multiple possible clustering.Finally,consistent quality function and majority vote are used to find the final clustering with high overall quality and non-overlapping from multiple possible clustering,and the service reputation is calculated based on the final clustering.The method fully considers the consistency relationship between the user classification of services and the final obtained reputation.The rationality and effectiveness of the method have been verified by experimental study.The experiments show that the method automatically calculates the high-quality reputation measurement results without inputting any parameters,and also improves the manipulation resistance ability of the reputation measurement method,so that users can make right services choice decision based on the reputation results.
刘艳丽,付晓东,岳昆,刘骊,冯勇,刘利军. 基于一致性聚类的在线服务信誉度量[J]. 小型微型计算机系统, 2020, 41(7): 1413-1420.
LIU Yan-li,FU Xiao-dong,YUE Kun,LIU Li,FENG Yong,LIU Li-jun. Online Service Reputation Measurement Based on Consensus Clustering. Journal of Chinese Computer Systems, 2020, 41(7): 1413-1420.
2020, Vol. 41 
