Making Political Decisions using Fuzzy Cognitive Maps: the FYROM crisis Athanasios K. Tsadiras1*, Ilias Kouskouvelis2 and Konstantinos G. Margarilis1 1Department of Applied Informatics 2 Department of International & European, Economic & Political Studies University of Macedonia, 54006 Thessaloniki, Greece Abstract. In this paper we use Fuzzy Cognitive Maps (FCMs), a well-established Arti - ficial Intelligence technique that incorporates ideas from Artificial Neural Networks and Fuzzy Logic, to create a dynamic model of the Former Yugoslavian Republic of Mace- donia (FYROM) Crisis in March 2001. FCMs create models as collections of concepts and the various causal relations that exist between these concepts. The decision capabili- ties of the FCM structure are examined and presented using a model developed based on the beliefs of a domain expert. The model is first examined statically using graph theory techniques to identify the vicious or the virtuous cycles of the decision process. The model is also examined dynamically thought simulations, in order to support political analysts and decision makers to their political decisions concerning the crisis. Scenarios arc introduced and predictions arc made by viewing dynamically the consequences of the corresponding actions. 1 Introduction to Fuzzy Cognitive Maps International Relations theory has long been concerned with Decision-Making, negotiations and crisis management [1], Cognitive Map (CM) models were introduced by Axelrod in the late 1970’s and were widely used for Political Analysis and Decision Making in International Relations [2], The structural and decision potentials of such models were studied and the explanation and prediction capabilities were identified [2,3], The introduction of Fuzzy Logic gave new representing capabilities to CMs and led to the development of Fuzzy Cog- nitive Maps by Kosko in the late 1980’s [4,5], The use of fuzzy logic allows the representa- tion both of the type (positive or negative) of the causal relationships that exist among the concepts of the model but also of die degree of die causal relationship. FCMs models are created as collections of concepts and die various causal relationships that exist between these concepts. The concepts are represented by nodes and the causal rela- tionships by directed arcs between the nodes. Each arc is accompanied by a weight that de- fines the degree of the causal relation between die two nodes. The sign of the weight deter- mines die positive or negative causal relation between the two concepts-nodes. An example of FCM is given in figure 1, showing die causal relationships diat were identified from Henry A. Kinssinger’s essay “Starting Out in die Direction of Middle East Peace” in Los Angeles Times 1982 and presented in [6]. In FCMs, although die degree of the causal relationships could be represented by a num- ber in die interval [-1,1], each concept, in a binary manner, could be eidier activated or not activated. In 1997, Certainty Neuron Fuzzy Cognitive Maps (CNFCMs) were introduced [7], ' Supported by a postdoctoral scholarship by the Greek National Scholarships Foundation (IKY). 501