一种环境敏感的可信模型
发布时间：2018-07-28 15:59
【摘要】：在选择网络中介服务的时候，信任和信誉系统的重要性越来越明显，信任和信誉系统的商业应用也越来越普级。可信模型作为各类信任和信誉系统的核心其基本理念是在某次交易结束以后，让参与者为对方在交易中的表现进行评判，然后用确定的评判生成信任和信誉的等级(可信值)，这个等级(可信值)可以帮助其它参与者决定是否与被评测者发生交易行为。目前，此领域内的研究者把主要精力放在如何将参与者的评价和评分科学有效的聚合成可信值，而对交易中的环境因素对交易行为的影响关注得不够，事实上环境对行为的影响力是巨大的，可信模型进行可信评估时应该将环境因素考虑在内。 本文提出了一种引入环境信息的可信评估模型。本模型首先通过引入倾向度的概念来刻画被评估实体在不同环境下表现出的不同行为倾向，进而提取出实体在不同环境下的行为模式。其次将实体行为模式中的各环境属性进行有效组合，生成衍生属性，行为模式扩展为扩展行为模式，以扩展模式来匹配交易环境，从而找出有特殊倾向性的实体，，准确定位最能适应当前交易环境的实体，提高模型应对客户复杂需求的能力。最后将与环境匹配成功的行为模式的倾向度与实体的一般可信值混调成环境敏感可信值。仿真实验证明，本模型能大幅度的提高用户的平均效率，具有良好的可用性。
[Abstract]:When choosing network intermediary service, the importance of trust and reputation system is becoming more and more obvious, and the commercial application of trust and reputation system is becoming more and more common. Trust model is the core of all kinds of trust and credit system, whose basic idea is to let the participants judge the performance of each other in the transaction after a transaction. Then a trust and credit rating (trustworthiness) is generated with a deterministic evaluation, which can help other participants decide whether or not to trade with the subject. At present, researchers in this field focus on how to effectively aggregate the participants' evaluation and scoring science into credible values, but pay less attention to the impact of environmental factors on trading behavior. As a matter of fact, the influence of environment on behavior is great, and the environmental factors should be taken into account in the credible assessment of trusted models. In this paper, a trusted assessment model with environmental information is proposed. In this model, the concept of tendency degree is introduced to describe the different behavioral tendencies of the evaluated entity in different environments, and then extract the behavior patterns of the entities in different environments. Secondly, each environment attribute of entity behavior pattern is combined effectively, and the derivative attribute is generated, and the behavior pattern is extended to extend behavior pattern to match the transaction environment, so as to find out the entity with special tendency. Accurately locate the entities that are most suitable for the current trading environment, and improve the ability of the model to deal with the complex needs of customers. Finally, the tendency of the successful behavior pattern matching with the environment is mixed with the general trusted value of the entity to form the environment-sensitive trusted value. The simulation results show that the model can greatly improve the average efficiency of users and has good usability.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.08
本文编号：2150811
[Abstract]:When choosing network intermediary service, the importance of trust and reputation system is becoming more and more obvious, and the commercial application of trust and reputation system is becoming more and more common. Trust model is the core of all kinds of trust and credit system, whose basic idea is to let the participants judge the performance of each other in the transaction after a transaction. Then a trust and credit rating (trustworthiness) is generated with a deterministic evaluation, which can help other participants decide whether or not to trade with the subject. At present, researchers in this field focus on how to effectively aggregate the participants' evaluation and scoring science into credible values, but pay less attention to the impact of environmental factors on trading behavior. As a matter of fact, the influence of environment on behavior is great, and the environmental factors should be taken into account in the credible assessment of trusted models. In this paper, a trusted assessment model with environmental information is proposed. In this model, the concept of tendency degree is introduced to describe the different behavioral tendencies of the evaluated entity in different environments, and then extract the behavior patterns of the entities in different environments. Secondly, each environment attribute of entity behavior pattern is combined effectively, and the derivative attribute is generated, and the behavior pattern is extended to extend behavior pattern to match the transaction environment, so as to find out the entity with special tendency. Accurately locate the entities that are most suitable for the current trading environment, and improve the ability of the model to deal with the complex needs of customers. Finally, the tendency of the successful behavior pattern matching with the environment is mixed with the general trusted value of the entity to form the environment-sensitive trusted value. The simulation results show that the model can greatly improve the average efficiency of users and has good usability.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.08
【参考文献】
相关期刊论文 前5条
1 蔡红云;田俊峰;李珍;何莉辉;;基于信任领域和评价可信度量的信任模型研究[J];计算机研究与发展;2011年11期
2 徐兰芳;胡怀飞;桑子夏;徐凤鸣;邹德清;;基于灰色系统理论的信誉报告机制[J];软件学报;2007年07期
3 王守信;张莉;李鹤松;;一种基于云模型的主观信任评价方法[J];软件学报;2010年06期
4 周林;刘仲英;;基于beta密度函数的信任模型[J];数学的实践与认识;2007年23期
5 田俊峰;吴志杰;;一种可信的云存储控制模型[J];小型微型计算机系统;2013年04期
本文编号：2150811
本文链接：https://www.wllwen.com/guanlilunwen/ydhl/2150811.html
