Nicos H. Mateou

The cyprus puzzle and the greek - turkish arms race: forecasting developments using genetically evolved fuzzy cognitive maps

The scope of this paper is to forecast the extent to which a settlement of the Cyprus issue may be possible given the decisions taken during the Copenhagen EU summit. It aims, in addition, at investigating the possibilities of improvement in Greek-Turkish relations which may lead, in turn, to reducing the arms race between the two countries. The paper uses a Genetically Evolved Certainty Neuron Fuzzy Cognitive Map algorithm to consider a number of scenarios examining the possible reactions of all sides involved in the Cyprus issue, namely Greece, Turkey, Cyprus, the Turkish-Cypriot community and the international environment. All simulation exercises suggest that the Greek and the Cypriot side should not necessarily rely on the decisions taken during the Copenhagen summit conference. The forecasts point out, in addition, that the optimism of the Greek government concerning the outlook of its relations with Turkey, and a subsequent reduction of the arms race against it, is far from being justified.

Multi-objective evolutionary fuzzy cognitive maps for decision support

This paper proposes an extension of genetically evolved fuzzy cognitive maps (GEFCMs) used for decision-making, aiming at increasing their reliability and overcoming its main weakness which lies with the recalculation of weights corresponding to more than one concept every time a new multiple scenario is introduced. A new evolutionary approach is proposed to support multi-objective decision-making based on the introduction of a dedicated genetic algorithm (GA), which is responsible for finding an optimal weight matrix that satisfies two or more activation levels among the participating concept nodes. This evolutionary methodology is very appealing since it offers the optimal solution without a problem-solving strategy once the requirements are defined

A new traversing and execution algorithm for Multilayered Fuzzy Cognitive Maps

This paper introduces a new algorithm for traversing and executing multilayered fuzzy cognitive maps (ML-FCMs) that aim to enhance this methodology, which is designed for handling complicated large scale problems. The methodology is based on the decomposition of the parameters of the problem under investigation into smaller quantities, organised in a hierarchical structure forming a multilayered FCM model. The present work aspires to eliminate the weaknesses of the existing ML-FCM algorithm, which reside in the way activation levels are calculated for those concepts decomposed into a set of parameters at lower layers in the map. The current algorithm calculates these levels by completing a full iteration cycle at the lower level thus losing the information produced between the iterative steps. We attempt to solve this problem by introducing the enhanced ML-FCM algorithm, (EML-FCM) which allows calculations in-between iterations and takes into consideration the change of activation levels in a more detailed form. The strong features of the proposed EML-FCM algorithm are presented and discussed, in addition to the provision of a comparison between the two algorithms.

Tree-Structured Multi-Layer Fuzzy Cognitive Maps for Modelling Large Scale, Complex Problems

This paper proposes tree structured multi-layer fuzzy cognitive maps for modelling large-scale and complex real world problems and supporting the decision making process. Large-scale problems are characterized by a large number of parameters, concepts, variables, nonlinearities and uncertainties that make their analysis and modelling a very difficult task. The objective of the proposed methodology is to give an alternative approach for dealing with the aforementioned difficulties, offering a new computational algorithm designed so as to support the creation of layers of parameters and variables describing the system under study, as well as the simulation of its evolution dynamics

Optimization in genetically evolved fuzzy cognitive maps supporting decision-making: the limit cycle case

This paper proposes an extension of genetically evolved fuzzy cognitive maps (GEFCMs) aiming at increasing their reliability by overcoming its weakness appearing in cases of a limit cycle behavior. FCMs use notions borrowed from artificial intelligence and neural networks to combine concepts and causal relationships, aimed at creating dynamic models that describe a given cognitive setting. The activation level of the nodes participating in an FCM model can be calculated using specific updating equations in a series of iterations.

Evolutionary Multilayered Fuzzy Cognitive Maps: A Hybrid System Design to Handle Large-Scale, Complex, Real-World Problems

This paper proposes an extension to multilayered fuzzy cognitive maps (ML-FCMs) and introduces a new methodology based on ML-FCMs aiming at enhancing their capabilities for scenario analysis and forecasting. The main issue here is the decomposition of the parameters into smaller, more manageable quantities, organised in a hierarchical structure forming a model, which consists of subsystems working together and supporting a central objective. The modelling of a particular large scale system is primarily represented by a main, central FCM, with distinct sub-models (layers) implemented also as FCMs and linked together in a hierarchical tree structure. The sub-models represent and implement (in computational terms) the decomposed parameters and variables of the system, thus offering the ability of isolating and studying critical parts of the system. The objective of the evolutionary multilayered FCM approach, as it is proposed in this work, is to improve the decision-making process of basic ML-FCMs by integrating a genetic algorithm (GA) for the production of a set of solutions in the form of new weight matrices for any targeted activation level throughout the multilayered structure

Fuzzy Influence Diagrams: An Alternative Approach to Decision Making Under Uncertainty

This paper proposes a strategic management methodology using influence diagrams to represent and model decision problems. While decision trees have been extensively used for this purpose and are still highly useful, fuzzy influence diagrams, a new representation for decision problems, outperform them in many respects. This paper describes influence diagrams, proposes their extension via fuzzy logic and demonstrates their use in crisis management and decision making. Simulation experiments using the proposed fuzzy influence diagrams indicate specific conditions and constraints that enable political analysts and decision makers to set targets and plan their actions on a prediction basis

Hybrid Fuzzy Cognitive Map Modeller: A Novel Software Tool for Decision Making

This paper presents the key concepts of an integrated software tool designed to contribute to the decision-making process in the field of crisis modelling and management. The tool relies on the use of Fuzzy Cognitive Maps (FCMs), which combine elements of fuzzy logic and neural networks to depict a cognitive scene of interacting concepts (nodes/levels) and their causal relationships (edges/weights). The proposed application provides the policy maker with the ability to input data from various domain experts in the form of activation levels and weight values, and to model the parameters of a certain environment. Simulations may then be performed and their results are presented and analysed for inference purposes and decision-making.

An Evolutionary Methodology to Eliminate the Limit Cycle Phenomenon in FCM-Based Models

This paper presents the dynamic behaviour of a hybrid system comprising fuzzy cognitive maps (FCM) and genetic algorithms, and focuses on the behaviour of the former under equilibrium at fixed points or limit cycle. More specifically, the theoretical background of both the equilibrium and limit cycle behaviours is examined and a new methodology for eliminating the limit cycle phenomenon is proposed. An evolutionary algorithm is designed based on the correction of the weight matrix, the origin of which is responsible for the creation of the limit cycle phenomenon. The system traces the presence of a limit cycle or chaotic behaviour and then searches the weight matrix using evolutionary techniques to identify which weight or group of weights are responsible for the cause of the limit cycle. This process continues until the system reaches equilibrium, a state which is necessary for the execution of simulations and the optimisation of various weights to meet a certain hypothetical scenarios in the context of decision-making

Facing Multicollinearity in a Fuzzy Cognitive Maps Environment Using Bias Nodes

This paper presents an improvement in the fuzzy cognitive maps (FCMs) theory used in modelling dynamical systems. The dynamic behaviour of a system focuses on a stable behaviour reaching equilibrium state at fixed points. Multicollinearity in FCMs, is examined as a phenomenon possibly present in any dynamic modelling. Cases in which the activation level of a concept is constantly driven towards +1 or -1 are investigated. In addition, concepts stabilizing at levels other than those reflecting the actual environment are also considered. These experiments are performed by selectively inserting bias nodes in the map so as to increase the reliability of the dynamic model. Following the theoretical background of FCMs and the mathematical formulation of bias nodes we demonstrate their use and contribution in terms of a case study that deals with the Cyprus issue in view of the Turkish full EU membership.