A New Model of Agent Self-Regulation Based on Profile Discovery in Social Networks Applied to the Ultimatum Game

This paper proposes a new model of interactions between agents who play the Ultimatum Game (UG) from the characteristics of knowledge discovery techniques and interactions in Social Networking Sites (SNS), more precisely on Twitter. To support the work, the authors present simulations using the UG with a spatial and evolutionary approach as well as technical knowledge discovery using SNS. With this we intend to find a more efficient way of interactions in UG, where failure is reduced in each round. For this purpose, the authors present here two new techniques that will be internalized in agents: the use of a historic reputation of the interactions between agents and, in certain periods of time, to perform the profile discovery profile of the agent offer in a general scope and their particular interactions with each agent.