Reputation systems can be used in online markets and communities in order to stimulate quality and good behaviour as well as to sanction poor quality and bad behaviour. The basic idea is to have a mechanism for rating services on various aspects, and a way of computing reputation scores based on the ratings from many different parties. By making the reputation scores public, such systems can assist parties in deciding whether or not to use a particular service. Reputation systems represent soft security mechanisms for social control. This article presents a type of reputation system based on the Dirichlet probability distribution which is a multinomial Bayesian probability distribution. Dirichlet reputation systems represent a generalisation of the binomial Beta reputation system. The multinomial aspect of Dirichlet reputation systems means that any set of discrete rating levels can be defined. This provides great flexibility and usability, as well as a sound basis for designing reputation systems.

Abstract. In many dynamic open systems, agents have to interact with one another to achieve their goals. Here, agents may be self-interested and when trusted to perform an action for another, may betray that trust by not performing the action as required. In addition, due to the size of such systems, agents will often interact with other agents with which they have little or no past experience. There is therefore a need to develop a model of trust and reputation that will ensure good interactions among software agents in large scale open systems. Against this background, we have developed TRAVOS (Trust and Reputation model for Agent-based Virtual OrganisationS) which models an agent’s trust in an interaction partner. Specifically, trust is calculated using probability theory taking account of past interactions between agents, and when there is a lack of personal experience between agents, the model draws upon reputation information gathered from third parties. In this latter case, we pay particular attention to handling the possibility that reputation information may be inaccurate. 1

Trust is an integral component in many kinds of human interaction, allowing people to act under uncertainty and with the risk of negative consequences. For example, exchanging money for a service, giving access to your property, and choosing between conflicting sources of information all may utilize some form of trust. In computer science, trust is a widelyused term whose definition differs among researchers and application areas. Trust is an essential component of the vision for the Semantic Web, where both new problems and new applications of trust are being studied. This paper gives an overview of existing trust research in computer science and the Semantic Web.

Abstract. QoS-based service selection mechanisms will play an essential role in service-oriented architectures, as e-Business applications want to use services that most accurately meet their requirements. Standard approaches in this field typically are based on the prediction of services’ performance from the quality advertised by providers as well as from feedback of users on the actual levels of QoS delivered to them. The key issue in this setting is to detect and deal with false ratings by dishonest providers and users, which has only received limited attention so far. In this paper, we present a new QoS-based semantic web service selection and ranking solution with the application of a trust and reputation management method to address this problem. We will give a formal description of our approach and validate it with experiments which demonstrate that our solution yields high-quality results under various realistic cheating behaviors. 1

This research aims to develop a model of trust and reputation that will ensure good interactions amongst software agents in large scale open systems. The following are key drivers for our model: (1) agents may be self-interested and may provide false accounts of experiences with other agents if it is beneficial for them to do so; (2) agents will need to interact with other agents with which they have little or no past experience. Against this background, we have developed TRAVOS (Trust and Reputation model for Agentbased Virtual OrganisationS) which models an agent’s trust in an interaction partner. Specifically, trust is calculated using probability theory taking account of past interactions between agents. When there is a lack of personal experience between agents, the model draws upon reputation information gathered from third parties. In this latter case, we pay particular attention to handling the possibility that reputation information may be inaccurate.

Reputation systems provide mechanisms to produce a metric encapsulating reputation for a given domain for each identity within the system. These systems seek to generate an accurate assessment in the face of various factors including but not limited to unprecedented community size and potentially adversarial environments. We focus on attacks and defense mechanisms in reputation systems. We present an analysis framework that allows for general decomposition of existing reputation systems. We classify attacks against reputation systems by identifying which system components and design choices are the target of attacks. We survey defense mechanisms employed by existing reputation systems. Finally, we analyze several landmark systems in the peer-to-peer domain, characterizing their individual strengths and weaknesses. Our work contributes to understanding 1) which design components of reputation systems are most vulnerable, 2) what are the most appropriate defense mechanisms and 3) how these defense mechanisms can be integrated into existing or future reputation systems to make them resilient to attacks.

Trust networks consist of transitive trust relationships between people, organisations and software agents connected through a medium for communication and interaction. By formalising a trust relationships, e.g. as reputation scores or as subjective trust measures, trust between parties within the community can be derived by analysing the trust paths linking the parties together. This article describes a method for trust network analysis using subjective logic (TNA-SL). It provides a simple notation for expressing transitive trust relationships, and defines a method for simplifying complex trust networks so that they can be expressed in a concise form and be computationally analysed. Trust measures are expressed as beliefs, and subjective logic is used to compute trust between arbitrary parties in the network. We show that TNA-SL is efficient, and illustrate possible applications with examples.

The term trust management suggests that trust can be managed, for example by creating trust, by assessing trustworthiness, or by determining optimal decisions based on specific levels of trust. The problem to date is that trust management in online environments is a diverse and ill defined discipline. In fact, the term trust management is being used with very different meanings in different contexts. This paper examines various approaches related to online activities where trust is relevant and where there is potential for trust management.

When transacting and interacting through open computer networks, traditional methods used in the physical world for establishing trust can no longer be used. Creating virtual network substitutes with which people, organisations and software agents can derive trust in other parties requires computerised analysis of the underlying trust networks. This article describes an approach to trust network analysis using subjective logic (TNA-SL), that consists of the three following elements. Firstly it uses a concise notation with which trust transitivity and parallel combination of trust paths can be expressed. Secondly it defines a method for simplifying complex trust networks so that they can be expressed in this concise form. Finally it allows trust measures to be expressed as beliefs, so that derived trust can be automatically and securely computed with subjective logic. We compare our approach with trust derivation algorithms that are based on normalisation such as PageRank and EigenTrust. We also provide a numerical example to illustrates how TNA-SL can be applied.

Abstract — The performance of distributed networks depends on collaboration among distributed entities. To enhance security in distributed networks, such as ad hoc networks, it is important to evaluate the trustworthiness of participating entities since trust is the major driving force for collaboration. In this paper, we present a framework to quantitatively measure trust, model trust propagation, and defend trust evaluation systems against malicious attacks. In particular, we address the fundamental understanding of trust, quantitative trust metrics, mathematical properties of trust, dynamic properties of trust, and trust models. The attacks against trust evaluation are identified and defense techniques are developed. The proposed trust evaluation system is employed in ad hoc networks for securing ad hoc routing and assisting malicious node detection. The implementation is fully distributed. Simulations show that the proposed system can significantly improve network throughput as well as effectively detect malicious behaviors in ad hoc networks. Further, extensive simulations are performed to illustrate various attacks and the effectiveness of the proposed defense techniques. I.