A Propitient Multi-agent System for Spontaneous Service Emergence in Pervasive Computing environments

In many pervasive applications, such as information sharing, a user is much more likely to communicate and interact with other users having similar interests. Therefore, an approach based on the creation of communities composed of users with similar preferences and interests is required to promote computer-based human interactions. In this paper, a reinforcement learning-based approach by mimicking the human immune system behavior for the implementation of pervasive applications is presented. In this model, services or user interests are represented by agents that can establish relationships based on affinities to create a spontaneous propitient multi-agent system. Affinity corresponds to the adequacy with which two services could bind to share common interest attributes. These affinities are adjusted or reinforced by users' satisfaction regarding their interaction and dynamic network condition changes. Simulation of this proposed service emergence model is also presented.