Social Knowledge for e-Governance

Project start: 2006/04/06 Project end: 2009/03/31 Support:

European Union

Project description:

APPLICATION BACKGROUND

The eRep project's aim was to build a theoretical framework of hypotheses about reputation, which are evaluated using human experiments, and later be used to implement agents according to humans' strategy. Using intelligent agents for trading resources in the Internet-of-Services is a research field between computer science and microeconomics, as both Grid computing (or Infrastructure-as-a-Service Cloud Computing) and software agents are about the design of distributed systems. Whereas the Grid community focused on the development of the "infrastructure and tools for secure and reliable resource sharing within dynamic and geographically distributed virtual organizations (VOs)", the agent community focused on the development of intelligent agents being able to act in uncertain and dynamic environments.

Using electronic institutions to secure grid environments can be implemented through agent technology. Our part in the eRep project was the extension of the electronic institution towards common interaction protocols (e.g. web services) and the simulation of market effects and reputation issues.

CONTRIBUTION TO THE MISSION OF MODUS

The concept of electronic institutions needs to be modelled and simulated, and it is both conceptually and methodologically inspired by researching in human institutions. In open multi-agent systems you have also autonomous entities that interact to achieve individual goals. The behaviour of these entities cannot be guaranteed. Therefore, and similar to what happens in human societies, you need mechanisms to guarantee the well working of the system in spite of the local behaviours. The use of an electronic institution that regulates the behaviour of agents the same way human institutions regulate the behaviour of people is one of this mechanisms that can be complemented by other mechanisms like, for example, the use of reputation.

RESULTS

The eRep project finished in 2009. Results have been applied in several dissertations to the usage of reputation in the Internet of services (König), strategies of cloud providers (Matros), negotiation protocols (Balke), insurability of cloud offerings (Streitberger), and SLA negotiations (Hudert).

Ongoing work researches into the reputation of App providers on app markets, trusting their privacy proposals, and in dealing with uncertain information on App markets and in specific coordination problems in the Healthcare domain.