AUTHENTICITY INCREASE OF DATA PROCESSING OF NON - STATIONARY OBJECTS BASED ON THE APPLICATION OF MULTI - CONTEXTUAL RECURRENT NEURAL NETWORK

The problem was formulated and methodical bases of construction of systems of the intellectual analysis of data based on hybrid model of identification for increase of the forecast reliability of casual modern processes were developed. The approaches were proposed and the principles, methods and algorithms of identification based on the combination of the possibilities of fuzzy sets, fuzzy logic and neural network were developed. The reliability of the developed technique was confirmed by the results of experimental studies.