Natural Immune System Response As Complexe Adaptive System Using Learning Fuzzy Cognitive Maps

In the Natural Immune Systems NIS, adaptive and emergent behaviors result from the behaviors of each cell and their interactions with other cells and environment. Modeling and Simulating NIS requires aggregating these cognitive interactions between the individual cells and the environment. In last years the Fuzzy Cognitive Maps (FCM) has been shown to be a convenient tool for modeling, controlling and simulating complex systems. In this paper, a new type of learning fuzzy cognitive maps (LFCM) have been proposed as an extension of traditional FCM for modeling complex adaptive system is described. Our approach is summarized in two major ideas: The first one is to increase the reinforcement learning capabilities of the FCM by using an adaptation of Q-learning technique and the second one is to foster diversity of concept's states within the FCM by adopting an IF-THEN rule based system. Through modeling and simulating response of natural immune system, we show the effectiveness of the proposed approach in modeling CASs.