Maksymova O. Automated system of identification examination based on fuzzy data model (exemplified by petrol).

Object - the process of quality control and identification of product samples. Goal - to increase the accuracy of the identification examination of product samples methods through the development of models, methods and software for information technology of identification examination. Methods: data analysis, pattern recognition theory in computational intelligence theory, fuzzy set theory, object-oriented programming. Theoretical and practical results: solved an actual scientific and practical task of developing models and methods for the identification of information technology examination and automate the process of its implementation (exemplified by petrol), thus improving the accuracy of the identification of product samples. The scientific novelty: for the first time developed and proved fuzzy model representation and processing of data for problem of identification examination, which allowed effectively to classify samples of products based on the modeling of examination cognitive processes and accumulated expert knowledge; improved method of constructing fuzzy classifier knowledge base through new methods of automatic construction of membership functions, which allowed to reduce the number of fuzzy rules of the knowledge base; has been further developed information technology of fuzzy classification based on the proposed fuzzy model, which takes into account the uncertainty of the problem identification examination, which allowed to solve the problem of categorical identification of the sample group and to increase the methods accuracy; has been further developed information technology of automated "soft" systems creation through the proposed organizational and functional structure of the automated system for identification examination, which allowed to improve the qualitative and quantitative characteristics of identification examination process. The degree of implementation: results of dissertation research implemented in a quality control laboratory of company "Parallel-M Ltd", Donetsk. Applications: Petrochemicals.