Search academic texts

Information × Registration Number 0212U005334, 0109U007673 , R & D reports Title Information technologіes of automation of pattern recognition and decision making for diagnosis under uncertainty based on hybrid fuzzy-logic, neural network, and multi-agent methods of computational intelligence popup.stage_title Head Subbotin Sergey Alexandrovich, Registration Date 24-07-2012 Organization Zaporizzhzhya State National University popup.description2 Object of research - the process of building of automated systems of technical diagnosis. Purpose of work - to increase accuracy, speed, and ability of control and level of automation of diagnosis and pattern recognition processes. In the development of intelligent diagnostic and recognition systems was developed in the work: a set of methods of synthesis of diagnostic and recognition models of dependencies on precedents that allows to automatically adjust weights and determine the structure of diagnostic models based on neuro-fuzzy networks, decision trees, associative rules and artificial immune systems which, in turn, saves the computer time and resources to build a model. It is also proposed quality models of sample and diagnostic neural model and sample forming methods, the use of which allows to solve the problem of building of automation systems of technical diagnostics. An experimental investigation of the properties and characteristics and recommendations was conducted by usage of the developed methods in practical problem solving of technical diagnostics. The results brought to practical realization in the form of computer programs. Results of research are recommended for use in solving problems of technical and biomedical diagnostics, pattern recognition by features and more. Product Description popup.authors Гофман Євгеній Олександрович Зайко Тетяна Анатоліївна Зайцев Сергій Олексійович Олійник Андрій Олександрович Олійник Олексій Олександрович Субботін Сергій Олександрович Федорченко Євген Миколайович popup.nrat_date 2020-04-02 Close