论文部分内容阅读
在具有动态性、虚拟计算和分布计算等特点的云计算环境中,如何采信云服务的可信度评价历史数据并由此挖掘最大的参考价值,是当前云服务的信任机制研究面临的一个关键问题.提出了一种云服务可信度评价的个性化融合方法,该方法从云服务的个性化特征分析入手,利用相似度评估函数,基于云服务用户集合构建特征共同体,接着引入证据理论,通过逐步求精迭代识别伪证据,过滤不可靠的可信度评价,然后,结合特征共同体的个性化加权系数,以一种新的相关证据融合公式实现可信度评价数据的融合.仿真实验和分析表明,本文方法具有良好的自适应性,能够为带有明显个性化特征且没有直接使用经验的用户提供具有较高价值的参考数据,从而提高云计算环境下服务可信度评价的准确度.
In the cloud computing environment with the characteristics of dynamic, virtual computing and distributed computing, how to adopt the credibility of cloud services to evaluate historical data and mine the maximum reference value is a key problem in the research of trust mechanism of cloud services at present This paper proposes a personalized fusion method for evaluating the trustworthiness of cloud services. This method starts with the analysis of personalized features of cloud services, builds a community of features based on cloud service users’ set, and then introduces evidence theory, Through iterative refinement, the false evidence is identified and the unreliable credibility is filtered, and then a new relevant evidence fusion formula is combined with the personalized weighted coefficient of the feature community to fuse the credibility evaluation data.The simulation experiment and The analysis shows that the proposed method has good adaptability and can provide reference data with high value to users with obvious personalization features and no direct experience, so as to improve the accuracy of service credibility evaluation in cloud computing environment .