系统仿真学报 ›› 2018, Vol. 30 ›› Issue (11): 4298-4312.doi: 10.16182/j.issn1004731x.joss.201811031

• 仿真应用工程 • 上一篇    下一篇

云模型属性加权聚类服务推荐信任度算法

王晋东1,2, 于智勇1,2, 张恒巍1,2, 方晨1,2   

  1. 1. 信息工程大学,河南 郑州 450001;
    2. 数学工程与先进计算国家重点实验室,河南 郑州 450001
  • 收稿日期:2016-12-02 修回日期:2017-03-14 发布日期:2019-01-04
  • 作者简介:王晋东(1960-), 男, 河南郑州, 硕士, 教授, 研究方向为云计算、信息安全; 于智勇(1992-), 男, 辽宁大连, 硕士, 研究方向为服务可信评估; 张恒巍(1984-), 男, 河南郑州, 博士, 研究方向为信息安全、云计算。
  • 基金资助:
    国家自然科学基金(61303074)

Service Recommended Trust Algorithm Based on Cloud Model Attributes Weighted Clustering

Wang Jindong1,2, Yu Zhiyong1,2, Zhang Hengwei1,2, Fang Chen1,2   

  1. 1. Zhengzhou Institute of Information Science and Technology, Zhengzhou 450001, China;
    2. State Key Laboratory of Mathematical Engineering and Advanced Computing, Zhengzhou 450001, China
  • Received:2016-12-02 Revised:2017-03-14 Published:2019-01-04

摘要: 云环境下存在不同质量的服务,如何选择可信度较高的服务是服务选择的关键问题.针对现有服务信任评估方法的不足,提出一种基于云模型服务属性加权的聚类方法,通过基于服务聚类的加权Pearson相关系数法计算用户信任评价相似度,结合用户服务选择指标权重进一步计算出用户相似度,从而选取最近邻居,通过服务推荐信任度算法,计算得到服务对于目标用户的推荐信任度。仿真实验表明,该算法能更加准确地计算出服务推荐信任度,有效满足用户在服务信任方面需求,为用户选取高质量可信服务提供有力的决策支持。

关键词: 云模型, 服务聚类, 用户相似度, 推荐信任度

Abstract: In cloud computing, there are cloud services of different qualities. How to choose credible and reliable service has become a key issue when users select services. Aiming at the shortcomings of the existing evaluation methods, a service clustering method based on attributes weighted cloud model is proposed. The users’ evaluation similarity with weighted Pearson correlation coefficient method based on service clustering is calculated. The users’ similarity combined with the index weights of users’ service selection is computed. The nearest neighbors are gotten. The recommendation trust of the services with the recommendation trust algorithm is obtained. Simulation results show that the proposed algorithm can calculate the service recommendation trust more accurately. It meets the demand of users in terms of trust of service and it has the practical significance in improving the quality of selected service.

Key words: cloud model, service clustering, user similarity, recommended trust

中图分类号: