Modeling and verification of trust and reputation systems

Abstract

Trust is a basic soft-security condition influencing interactive and\ud cooperative behaviors in online communities. Several systems and models have\ud been proposed to enforce and investigate the role of trust in the process of\ud favoring successful cooperations while minimizing selfishness and failure.\ud However, the analysis of their effectiveness and efficiency is a challenging\ud issue. This paper provides a formal approach to the design and verification of\ud trust infrastructures used in the setting of software architectures and computer\ud networks supporting online communities. The proposed framework encompasses a\ud process calculus of concurrent systems, a temporal logic for trust, and model\ud checking techniques. Both functional and quantitative aspects can be modeled and\ud analyzed, while several types of trust models can be integrated

A Formal Framework for Modeling Trust and Reputation in Collective Adaptive Systems

“…Each club is defined as a group including one agent of type CDSR, some consumer, and several producers offering the service that characterizes the club. For instance, given two fixed clubs G 1 and G 2 , the process term Cons could be defined as follows (the summation symbol ∑ is used to generalize the choice operator): (1).Cons + obs(−1).Cons) + receive denial ij .obs(−1).Cons)…”

“…The formal framework proposed in this paper can be used as a basis for the verification of distributed trust systems. For this purpose, in [1], a model checking based approach is defined that relies on a trust temporal logic, called TTL, which is defined for the verification of TLTS-like models and, e.g., can be mapped to the logic UCTL [4]. Here, we specify the atomic statements of such a logic, which depend on the representation of trust information in our calculus, while the logical and temporal operators can be found in [1].…”

“…For this purpose, in [1], a model checking based approach is defined that relies on a trust temporal logic, called TTL, which is defined for the verification of TLTS-like models and, e.g., can be mapped to the logic UCTL [4]. Here, we specify the atomic statements of such a logic, which depend on the representation of trust information in our calculus, while the logical and temporal operators can be found in [1]. Similarly as for other logics merging action/state based predicates, atomic formulas include actions labeling TLTS transitions and state-based trust predicates:…”

“…Classical verification techniques, like model checking, have demonstrated their adequacy in the validation process of systems with respect to properties like safety, reliability, security, and performance. On the other hand, they have not received the same attention in the setting of trust and reputation (see, e.g., [1] and the references therein). To cite few representative examples in the setting of model checking based analysis, Reith et al [25] verify delegation mechanisms in access control, which can be viewed as a form of trust management, while He et al [12] apply the same approach to the verification of chains of trust.…”

On Contract Design for Incentivizing Users in Cooperative Content Delivery With Adverse Selection

“…In heterogenous cooperative wireless networks, the proper design of incentive mechanisms is crucial in order to motivate users to participate in collaborations such as for offloading traffic. Broadly speaking, the different forms of incentives that can be used to encourage users in a network to cooperate can be classified as (a) monetary [1], [23] and (b) non-monetary incentives [24], [25] which we review next.…”