Fuzzy sets and the theory of neuronal group selection for the problem of classification

Usually the procedure of classification is strongly associated with the pattern recognition problem. Within this problem the classification procedure is applied to distinguish visual images (for example, letters), sound and tactile patterns, etc. One of the difficulties we can meet on this way is the impossibility to give a comprehensive formal description to the classes we wish to distinguish. The proposed mechanism of classification, based on the Edelmen´s neuronal group selection theory, can to a certain extent eliminate such sort of difficulties. The key feature of this mechanism lies in the way it creates structures that provide recognition. These structures are formed as a result of selection of neuronal, compositions (groups), the process that adopts mechanisms of population genetics and therefore makes the created structures adequate to hardly formalized impacts of environment like the natural evolutionary process does. The possible area of application of the proposed mechanism can be extended beyond the problem of pattern recognition. Such a mechanism can be used when simulating the other procedures peculiar to the human mind, for example, procedure of interpretation