Emergence and Evolution of Agent-Based Referral Networks

Abstract

Numerous studies have shown that interpersonal communication acts as an important channel for gathering information. But if we wish to rely oninterpersonal communication, we still need to figure out how to determinethe right person to ask. Usually we cannot find the potential expert(s) directly, and we need some assistance from our friends or friends' friends to locate them. The phenomenon of Six Degrees of Separation indicates that it is possible to use some intelligent software agents, who can interpret the links between people and follow only therelevant one, to find the desired experts quickly. A computational model of agent-based referral networks was proposed to assist and simplify the users to find potential experts for a specified topic in a person-to-person social network, in which each user is assigned a softwareagent, and software agents help automate the process by a series of "referral chains''. Unlike most previous approaches, our architecture is fully distributed and includes agents who preserve the privacy and autonomy oftheir users. These agents learn models of each other in terms of expertise(ability to produce correct domain answers), and sociability (ability to produce accurate referrals). We study this framework experimentally to see the effects that the different variables have on each other and the efficiencyof the referral networks. Furthermore, a social mechanism of reputation management was proposed tohelp agents (users) avoid interaction with undesirable participants inthe referral networks. The mathematical theory of evidence is used torepresent and propagate the reputation information in a referral network.Our approach adjusts the ratings of agents based on their observations as well the testimony from others. Moreover, we conducted several experimentsto study the reputation management in different settings. Social mechanismsare even more important when some centralized reputation managementmechanisms, i.e., trusted third parties, are not available. Our specific approach to reputation management leads to a decentralized society in which agents help each other weed out undesirable players.