Jordi Sabater

Review on Computational Trust and Reputation Models

The scientific research in the area of computational mechanisms for trust and reputation in virtual societies is a recent discipline oriented to increase the reliability and performance of electronic communities. Computer science has moved from the paradigm of isolated machines to the paradigm of networks and distributed computing. Likewise, artificial intelligence is quickly moving from the paradigm of isolated and non-situated intelligence to the paradigm of situated, social and collective intelligence. The new paradigm of the so called intelligent or autonomous agents and multi-agent systems (MAS) together with the spectacular emergence of the information society technologies (specially reflected by the popularization of electronic commerce) are responsible for the increasing interest on trust and reputation mechanisms applied to electronic societies. This review wants to offer a panoramic view on current computational trust and reputation models.

Reputation and social network analysis in multi-agent systems

The use of previous direct interactions is probably the best way to calculate a reputation but, unfortunately this information is not always available. This is especially true in large multi-agent systems where interaction is scarce. In this paper we present a reputation system that takes advantage, among other things, of social relations between agents to overcome this problem.

Social ReGreT, a reputation model based on social relations

Social network analysis has been used for a long time by sociologist as a mechanism to infer and explain social behaviours. Analysing social relations between individuals in a society using social network analysis techniques allows to identify a lot of features of that society that can be very valuable in an e-commerce environment. In this paper we describe how these techniques can be used to improve a reputation system and also how this system can be integrated with a negotiation model to increase the succes of negotiations.

A specification of the Agent Reputation and Trust (ART) testbed: experimentation and competition for trust in agent societies

A diverse collection of trust-modeling algorithms for multi-agent systems has been developed in recent years, resulting in significant breadth-wise growth without unified direction or benchmarks. Based on enthusiastic response from the agent trust community, the Agent Reputation and Trust (ART) Testbed initiative has been launched, charged with the task of establishing a testbed for agent trust- and reputation-related technologies. This testbed serves in two roles: (1) as a competition forum in which researchers can compare their technologies against objective metrics, and (2) as a suite of tools with flexible parameters, allowing researchers to perform customizable, easily-repeatable experiments. This paper first enumerates trust research objectives to be addressed in the testbed and desirable testbed characteristics, then presents a competition testbed specification that is justified according to these requirements. In the testbeds artwork appraisal domain, agents, who valuate paintings for clients, may gather opinions from other agents to produce accurate appraisals. The testbeds implementation architecture is discussed briefly, as well.

Trust evaluation through relationship analysis

Current mechanisms for evaluating the trustworthiness of an agent within an electronic marketplace depend either on using a history of interactions or on recommendations from other agents. In the first case, these requirements limit what an agent with no prior interaction history can do. In the second case, they transform the problem into one of trusting the recommending agent. However, these mechanisms do not consider the relationships between agents that arise through interactions (such as buying or selling) or through overarching organisational structures (such as hierarchical or flat), which can also aid in evaluating trustworthiness. In response, this paper outlines a method that enables agents to evaluate the trustworthiness of their counterparts, based solely on an analysis of such relationships. Specifically, relationships are identified using a generic technique in conjunction with a basic model for agent-based marketplaces. They are then interpreted through a trust model that enables the inference of trust valuations based on the different types of relationships. In this way, we provide a further component for a trust evaluation model that addresses some of the limitations of existing work.

Engineering Executable Agents using Multi‐context Systems

In the area of agent-based computing there are many proposals for specific system architectures, and a number of proposals for general approaches to building agents. As yet, however, there are comparatively few attempts to relate these together, and even fewer attempts to provide methodologies which relate designs to architectures and then to executable agents. This paper provides a first attempt to address this shortcoming. We propose a general method of specifying logic-based agents, which is based on the use of multi-context systems, and give examples of its use. The resulting specifications can be directly executed, and we discuss an implementation which makes this direct execution possible.

Rapid Prototyping of Large Multi-Agent Systems Through Logic Programming

Prototyping is a valuable technique to help software engineers explore the design space while gaining insight on the dynamics of the system. In this paper, we describe a method for rapidly building prototypes of large multi-agent systems using logic programming. Our method advocates the use of a description of all permitted interactions among the components of the system, that is, the protocol, as the starting specification. The protocol is represented in a way that allows us to automatically check for desirable properties of the system to be built. We then employ the same specification to synthesise agents that will correctly follow the protocol. These synthesised agents are simple logic programs that engineers can further customise into more sophisticated software. Our choice of agents as logic programs allows us to provide semi-automatic support for the customisation activity. In our method, a prototype is a protocol with a set of synthesised and customised agents. Executing the prototype amounts to having these agents enact the protocol. We have implemented and described a distributed platform to simulate prototypes.

EVALUATING THE ReGreT SYSTEM

The ReGreT system is a trust and reputation mechanism that uses, among other things, social information to improve the calculation of trust and reputation measures. Using a framework called SuppWorld designed to test these kind of complex models, we present a set of experiments that evaluate different features of the ReGreT system in several scenarios.Jordi Sabater enjoys a Marie Curie Intra-European Fellowship (European 6th framework program) contract No MEIF-CT-2003-500573

Skeleton-based agent development for electronic institutions

In this paper we describe an approach for semi-automatic agent development. We focus on the scenario in which agents are designed to follow an electronic institution, a formalism to specify open agent organisations. In our approach, an initial design pattern is automatically extracted from a given electronic institution. This pattern is then offered to programmers willing to develop agents to perform in the electronic institution. Our approach supports developers when modifying the initial simple design pattern into more sophisticated programs.

Agent Specification Using Multi-context Systems

In the area of agent-based computing there are manyproposals for specific system architectures, and a number of proposals for general approaches to building agents. As yet, however, there are comparatively few attempts to relate these together, and even fewer attempts to provide methodologies which relate designs to architectures and then to executable agents. This paper discusses an attempt we have made to address this shortcoming, describing a general method of defining architectures for logic-based agents which can be directlye xecuted. Our approach is based upon the use of multi-context systems and we illustrate its use through examples of the specification of some simple agents.