Raimundas Jasinevicius

Fuzzy expert maps for risk management systems

After widely known implementations of perfect ideas, expressed by researchers R. Axelrod, L. Zadeh and B. Kosko, the fuzzy cognitive map (FCM) became an important tool for decision makers in various practical areas, such as business evaluation, risk management, international policy making, medical diagnostics and others. The principle possibility of FCM involvement into a regional oceanic modeling system (ROMS) was presented during the US/EU Baltic International Symposium in Klaipeda in May 2006. Todaypsilas experience permits to extend the concept of FCM nodes, including additional fuzzy expert knowledge and enriching the representation of real situations under consideration. This paper presents a systematic approach, based on the authorspsila research, to the idea of FCM extension and its transformation into a rule-based fuzzy expert map (FEM), presents a simplified fragment of FEM use for a port security system, and discusses further perspectives of developing new tools for decision makers in risk management systems in general.

On Fundamentals of Global Systems Control Science (GSCS)

Globalization leads us towards dealing with very complex systems that consist of evolving, overlapping, and interacting “socio-technical fabrics”. An existing general systems control theory cannot cope with problems occurring in such systems. This chapter is, first of all, an attempt to present an entirely new approach to the adequacy of system model and reality, based on a causal correspondence between information and knowledge obtained from a reality and its model. Secondly, the chapter suggests two possible control loops: one is meant to improve the model and another is the way to attain a certain planned goal to be reached by our reality. Four doctrines are presented as the basic principles of general fuzzy systems control theory (GFSCT) aiming to deal with the real fuzzy systems operating and functioning in a multiple space-time coordinate system. The minimization of a certain potential V-function is considered as a universal principle for existence of each system in the real world. Moreover, decentralized stochastic control is proposed to improve our reality and guarantee its lifetime unlimited behavior with a proper degree of certainty and space-time stability.

Fuzzy Hyperinference-Based Pattern Recognition

The paper presents a new approach to the problem of pattern recognition. First of all, here is emphasized that the problem itself is fuzzy enough. Later three following novelties of the approach are disclosed: 1) the rule-based fuzzy inference, concerning the measure of patterns’ similarity, is enriched by an idea of hyperinference; 2) a description of the main pattern recognition process is based on Takagi-Sugeno (T-S) reasoning procedure and 3) rule weights in T-S procedure are defined, solving special linear or piecewise linear programming problem (LPP or PWLPP), constructed according to the certain fuzzy experts’ information. The proposed approach was used successfully for recognition of healthy people and those who suffer from certain illness (for example, an atherosclerosis). The classification was performed according to person’s clinical posturograms (stabilograms). At the end of this paper experimental results are presented as well as acknowledgement to all anonymous participants of the experiments.

Fuzzy regularized generalized eigenvalue classifier with a novel membership function

Supervised classification of data affected by noise or error, with unknown probability distribution, is a challenging task. To this extend, we propose the Fuzzy Regularized Eigenvalue Classifier, based on a recent technique to classify data in two or more classes. We compare the execution time and accuracy of the classifier with other de facto standard methods. With the adoption of a novel membership function, the classifier is capable to produce more accurate models that well compare with results obtained by other methods, and fuzzy weighting functions, on benchmark datasets.

Fuzzy expert maps: The new approach

The paper presents a new approach to fuzzy knowledge management and provides fuzzy expert maps (FEM) as a tool. FEM is a systematic extension for a well-known paradigm of fuzzy cognitive maps combined with open fuzzy control systems. Transparent examples from international politics are presented to illustrate the entire extension chain. Recommendations for further research are given as well.

Methods for Smart Home Environment‘s Intellectualization: The Comparative Analysis

The difference between smart home (or smart agent) and intellectual- ized smart home (or intellectual smart agent) is delivered, and the framework for their modelling is described. Question of a sophisticated adequacy evalua- tion is raised to compare the reality and its model. The proposed approach was used to evaluate four different decision making technologies implemented in the models of the intellectualized smart home environment, and the comparison of those experimental technologies prepared by expert is presented.

Statistical Evaluation of Four Technologies used for Intelectualization of Smart Home Environment

This paper addresses the issues of decision-making methods and their usage capabilities for intelligent control based on resident’s habits. Learning from the behaviour of the resident is essential for the system to adapt and provide intelligent control based on behaviour patterns. Different homes have different conditions and habits which have to be taken into account for the intelligent system to be useful. However, even deeply ingrained habits are subject to change over time. Therefore, an intelligent system has to respond to changing and diverse environment. Various decision-making methods have the potential of a number of benefits in providing intelligent control for the Smart home systems. In this paper, concurrent decision-making methods, including Artificial Neural Networks, Fuzzy Logic, Linear Programing and Bayesian, are employed with particular algorithms in order to provide control based on resident’s habits. These approaches are tested and compared within experimental scenarios for intelligent lightning control. DOI: http://dx.doi.org/10.5755/j01.itc.44.3.11965

Why today's systems theory can't cope with global environmental or marine systems catastrophes and crises?

This presentation is based on the first reaction of the world famous systems theory scientists and researchers to the global events on our earth (starting with the contemporary banking crisis and finishing with the global environmental and marine systems catastrophes, which recently have shaken or are still threatening to shake the whole world). Comments on the events dealt with political, social and systems control theory aspects. The global management of world marine ecosystem, economy and social system was considered as SoC — the system of communities aiming a maximal profit instead of rational efficiency. Such an approach was based on a certain institutionalization of a global greed. On the other hand the environmental and marine systems are presented as a stochastic set of interdependent world-vide information and control systems considered as SoS — the system of systems, having no instruments (neither fundamental theory nor tools) able to cope with the current critical situation. The old and perfectly developed control theory with its well-known stability criteria is not adequate to the phenomenon of the contemporary worlds realities. New approach is proposed, adequate to the global arrangements and interdependence of actors in the contemporary world system of systems: 1) consideration of a new type of information — fuzzy knowledge-based data; 2) development of an extended independent space-time world coordinate system; 3) introduction and inclusion into the theory of a very important feature, such as anisotropy of our worlds structure. These three aspects up till now were neglected by the general systems theory, and this negligence coused societys scientific impotence in the presence of the crises. So, a new branch of general systems theory is under development now; its name — GFSCT — general fuzzy systems control theory, and it is taking into account all three approaches mentioned above. In this presentation is emphasized also that the new theory, as well as worlds community political will, is crucial for our future.

Dynamic SWOT analysis as a tool for ROMS

Regional oceanic modeling systems (ROMS) as multi-purpose and multi-disciplinary entities are too complicated to be based on strictly quantitative mathematical descriptions. A verbal qualitative presentation of the regional oceanic ecosystem is inevitable. At the moment well known SWOT (strengths/weaknesses/opportunities/threats) analysis mechanism is highly recommended for the evaluation of ecosystems behavior. Unfortunately such an approach permits us to cope with the problems of their static. Thorough modeling, monitoring and prediction require certain level of dynamics. The paper proposes the joint SWOT analysis and fuzzy cognitive maps (FCM) approach for the purpose mentioned above and shows that such a hybridization enables us to enrich ROMS with the new tool.

Solution of the Augmenting Sequence of Linear Programming Problems as a Tool for the Intellectual Home Environment’s Self-Training

This work presents the solution based on the augmenting sequence of linear programming problems (LPP) as a tool for intellectualizing home environment. The proposed solution empowers the intelligent decision making procedure which can be applied to various intelligent control applications. The augmenting self-training procedure based on LPP approach is presented as well, which allows making reasonable decisions having only limited data about the controlled environment. The method permits retraining the decision making system when new data is available. As a proof of concept, this solution is applied to intelligent light control application. The obtained simulation results show the method’s capability in making reasonable decisions according to users preferences. DOI: http://dx.doi.org/10.5755/j01.itc.44.2.10235