摘要
针对网格服务信任评估中难以鉴别恶意评估的问题,提出一种基于赔偿的网格服务信任评估模型。模型中,引入市场机制,评估双方对评估行为提出期望收益并承诺赔偿价格,双方目的是最大化各自收益。借助信号博弈理论,通过对模型进行分析和求解,得出评估双方基于赔偿的评估交易策略,策略满足完美贝叶斯分离均衡。评估双方可根据策略选择自己的交易行为,从而使各自收益最大化。理论分析和仿真结果表明:该模型能使信任评估者主动摒弃恶意评估,服务提供者可根据赔偿价格有效地判断评估的可信度,主动选择评估对象,与现有模型相比,在精确信任评估的准确性、服务的可靠性及简化计算、降低通信开销等方面有较大的提高。
It is difficult to distinguish the cheating and other malicious behaviours in the grid service trust evaluation. A trust evaluation model for grid service based on compensation is proposed. With market mechanism, service provider and demander present the expected profit and compensate price for their behaviours to maximize their own profit. The game theory is applied to balance the expected profit and compensate price. Based on the game theory, evaluation trade strategy achieves perfect Bayesian equilibrium. Service provider and demander can choose their own trade strategy to maximize profit. Theoretical analysis and simulation results show that the service demanders can abandon the cheating motivation voluntarily, while the service providers can make right decision to choose trust evaluation. Compared with the existing model, the precision and security of the trust evaluation is enhanced greatly, while the calculating and communication cost is reduced remarkably.
出处
《重庆大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2010年第6期121-127,共7页
Journal of Chongqing University
基金
国家自然科学基金资助项目(70671011)
国家863计划资助项目(2007AA04Z1B1)
关键词
网格服务
赔偿
博弈
信任评估
grid service
compensation
game
trust evaluation