基于可能世界的网构软件模型及可信性研究  

Internetware model based on possible worlds and research of creditability

作　　者：刘超[1] 王文杰[1] 

机构地区：[1]中国科学院研究生院(本部),北京100049

出　　处：《计算机应用研究》2010年第12期4604-4607,4611,共5页Application Research of Computers

基　　金：国家"863"计划资助项目(2007AA01Z132)

摘　　要：为了解决网构软件的应用建模中发生的主观逻辑冲突问题,首先从认识论入手,提出了影子实体的概念,将逻辑冲突归结为实体内涵的不一致所致;为了解决影子实体的自身安全问题,引入了可能世界理论,给出了一种可能世界的语义学解释,并在可能世界的基础之上构建了软件实体交互模型。然后,结合形式概念分析技术,提出了信任格的概念。最后,讨论了信任模型的几点应用,证明了资源分配公平存在性定理,在资源分配问题上为阿罗不可能定理提供了新的解决方案,进而表明主观逻辑冲突问题是可解的。To overcome the problem of logistic conflict in the application modeling of internetware,first of all,the paper presented a concept of shadow entities,and attributed the conflict to the difference of entity＇s connotation. To deal with the problems of shadow entities＇security,introduced the possible worlds theory,and based on the theory of possible worlds,built an interactive model for internetware. Then,following the theory of formal concept analysis,proposed a concept of trust lattice. Finally,discussed some application under the subjective trust model and proved the fair resource Allocation＇s possible theo-rem. For the problem of resource allocation,provided a new solution for Arrow＇s impossibility theorem and it also showed that the problem of conflict was solvable.

关 键 词：网构软件 可能世界 可信建模 影子实体 信任格 资源分配公平存在性定理 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]