Research object - processes of fuzzy models adaptation in uncertainty conditions. Research target - development of methods for fuzzy models adaptation using artificial immune systems, which allowed to increase efficiency of information processing in uncertainty conditions and to decrease influence of subjective expert estimation. Methods of research - theory of fuzzy sets, theory of artificial neural networks, theory of artificial immune systems which allowed the development and adaptation of fuzzy models; probability theory, mathematics statistics theory, theory of Markov chains which allowed to prove the convergence of the designed immune algorithms; theory of parallel computing which allowed to rise efficiency of immune algorithm of fuzzy neural network adaptation. Theoretical and practical results - methods of fuzzy models adaptation and fuzzy neural network adaptation based on artificial immune systems using experimental data. Scientific novelty - the methods of structural and parametric adaptation of fuzzy models adaptation based on artificial immune systems, are proposed for the first time, which allows to simplify fuzzy inference models and increase the precision; the method of adaptation of fuzzy neural network based on artificial immune systems to simultaneous adjustment of its structure and parameters, is proposed for the first time, which allows to simplify the network; the model of coding of adjusted parameters of fuzzy neural networks in the form of the structured adaptive multiantibody in which parameters are divided into independent parts, and the size of a multiantibody is not fixed, is proposed for the first time, which allows to carry out simultaneous adjustment of parameters of a network and to reduce amount of neurons in the latent layers of a network; method of convergence estimation of the immune algorithms, characterized by the representation of action of the immune mutation and edit operators in the form of a transition matrix and use of real coding of antibodies, is improved; methods of mutation and cloning antibodies, are improved, which allowed to increase the convergence velocity of immune algorithms of fuzzy models adaptation. Degree of implementation -research results are used in the Institute of dermatology and venerology of Medical Sciences Academy of Ukraine (act 31.08.09), in the GP NITIP (act 14.08.09), and also are used in the learning process of Kharkov national university of radioelectronics which is confirmed by the act of 16.10.09. The scope of use - in organizations that deal with similar problems of developing methods for fuzzy models adaptation, in the areas of information technology, finance, medicine, biology, ecology, and in the educational process in the preparation of specialists in the areas of intellectual information processing.
