Utilising social recommendation for decision-making in distributed multi-agent systems

Open multi-agent systems are typically formed from heterogeneous peers operating in a decentralised manner. Hence, their constituent agents must evaluate possible actions and opportunities based on local, subjective knowledge. When agents have insufficient personal experience, they may inevitably rely on their social connections to act as a source of relevant information or recommendations. We describe an agent-mediated electronic market for investigating social interaction within the context of evolving heterogeneous distributed networks. In our scenario, consumers look for appropriate services and this service choice is informed via peer recommendations. We define two alternative algorithms for selecting peers based on perceived similarity and we evaluate them on their ability to organise an overlay network such that it acts as a passive filter, tailoring the information that agents use to select services in the market. We use this scenario to explore the link between the peer selection algorithms and the emergent network topologies, as well as the impact of the peer selection algorithm on the agents' performance in choosing services based on peer recommendations. Our simulation results demonstrate a qualitative difference in the behaviour of the algorithms, with optimal algorithm selection relying on information regarding the preferences of the wider population of agents.