Evolutionary Fuzzy Cognitive Maps: A Hybrid System for Crisis Management and Political Decision Making

This paper proposes an extension of Certainty Neuron Fuzzy Cognitive Maps (CNFCMs) used in crisis management and decision-making, aiming at increasing their reliability. The objective of the Genetically Evolved Certainty Neuron Fuzzy Cognitive Map (GECNFCM) as it is introduced here is to overcome the main weakness of CNFCMs, which lies with the recalculation of the weights corresponding to each concept every time a new strategy is adopted. The problem is overcome through the introduction of a Genetic Algorithm (GA), which produces a set of solutions and new weights following a strategy change. The GA concepts are very appealing since they offer the optimal solution without a problem-solving strategy, once the requirements are defined. It is interesting to point out that the hybrid approach is reflected in both the implementation of the GA and in the methodology applied for solving the problem. In fact, the reasoning behind this hybrid methodology is to use it for obtaining the optimal values of the weights corresponding to the variables of the model rather than the optimal values of the variables themselves.