Recommendations Management in a Trust Model in Service Oriented Architecture
In Service-Oriented Architecture(SOA), users and services are often required to interact with unknown entities.With the increasing uncertainty and risk, trust models have emerged as an important risk management mechanism in such online interactions. These trust models differ in aggregations of recommendations. This paper presents a trust model of recommendations management based on Bayesian estimation and the neural network. By using Bayesian estimation the Dirichlet distribution, our proposal integrates the subjective prior knowledge and the actual experience into the estimation of trust. By a novel application of neural network, our proposal evaluates multiple recommendations of various trust standards with and without deceptions. We describe simulation experiments to illustrate the effectiveness and robustness of the methods.
trust model recommendation bayesian estimation,neural network
LI Lifen ZHAO Huaiyu
Department of Computer Science and Technology North China Electric Power University No.619 Yonghuabe Department of Electron and Communication Engineering North China Electric Power University No.619 Yo
国际会议
武汉
英文
570-573
2007-07-25(万方平台首次上网日期,不代表论文的发表时间)