一种集成直觉模糊信息的激励自适应信任模型

Incentive Adaptive Trust Model Based on Integrated Intuitionistic Fuzzy Information

针对现有信任机制不能很好表达信任的不确定性,且缺乏有效处理分布式网络中存在的不诚实推荐和策略性欺骗等问题,该文提出一种集成直觉模糊信息的自适应信任评估模型。为了激励提供可信服务的实体,惩罚不良行为实体,该模型结合服务持续性强度和时间衰减因子,计算实体直接信任直觉模糊数。同时利用实体的推荐可信度和行为一致性过滤不诚实推荐行为。除此之外,为了克服综合信任聚合计算中权重确定较主观的问题,给出了一种新的自适应权重分配方法。仿真实验表明,该模型在提高实体服务积极性和抑制恶意欺诈行为中,都有较好的适应性和有效性。

Existing trust models can not effectively express the uncertainty of trust relationship and deal with such issues as dishonest feedbacks and strategic frauds from malicious entities in the distributed network, an adaptive trust model based on aggregating Intuitionistic Fuzzy Information（IFI） is proposed. Firstly, in order to incentivize entities providing trustworthy service and punish entities taking along malicious behavior, an approach on aggregating IFI is constructed to compute the direct trust intuitionistic fuzzy numbers which contain the latest permanence factor and the time decay factor. Then, the recommendation credibility and uniformity are defined to detect dishonest recommendation. Subsequent, an adaptive weighted approach is developed to avoid distributing the weights of direct and indirect trust subjectively. The simulation experiments demonstrate that the proposed model not only is robust on malicious attacks, but also has better adaptability and effectiveness.