P2P reputation management through social networking

Abstract

Reputation systems offer a viable solution to the problem of risk reduction in online communities, in situation in which other mechanism such as litigation or security cannot help. Building on the assumption that its participating entities engage in repeated interactions, a reputation system can either signal what happened in the past or aggregate the past feedback in such a way as to influence the future actions of the concerned entity. In the former case, the concerned entity's behavior is seen as static, while the sent signal is expected to be indicative of the entity's future actions. In the latter case, behavior is dynamic in the sense that the entity can adjust it given the observed feedback, while the purpose of the reputation system is to induce adjustments according to the designer's needs. In this chapter, we discuss these two classes of solutions in detail. In particular, we investigate how they apply to P2P networks, what additional problems and difficulties the P2P environment introduces and what scalable solutions to these problems the current research offers. © 2010 Springer Science+Business Media, LLC.