摘要： 当前的信任模型缺乏足够的信任评价控制机制，严重影响信任模型在网格等开放计算环境中的应用。基于传统的信任评价模型，提出信任评价控制模型。该模型采用插值技术，通过服务水平估计，有效过滤和调谐用户评价，对用户提出激励处罚机制。仿真结果表明，该模型优于现有的控制模型，能够较好地解决网格实体信任评价可靠性问题，降低恶意评价对信任评价模型的影响。

关键词: 网格环境, 评价控制, 插值技术, 激励

Abstract: Current trust models lack sufficient control mechanism for trust evaluation. It seriously influences the application of trust models under open computing environment such as grid. Based on traditional trust evaluation model, a trust model with evaluation control is presented, in which interpolation technique is introduced and corresponding incentive and punishment mechanism are put forward. With the estimation of service quality, the model effectively filters and tunes the user’s rating. Simulation shows that the model outperforms the existing control model, and can well solve the reliable trust evaluation problem of grid entity, and can decrease the influence of malicious ratings on trust evaluation models.

Key words: grid environment, evaluation control, interpolation technique, incentive

中图分类号: 

• TP393