Recognition and classification problems with standard training information

Abstract A SURVEY is given of the results on recognition and classification problems obtained by the members of the Laboratory of Recognition Problems of the Computing Centre of the Academy of Sciences of the USSR in the period 1969–1980. Individual recognition algorithms, families (models)of recognition algorithms, and also linear, algebraic, algebraic of degree κ closures of families of recognition algorithms are considered. Algorithms for solving taxonomy problems (training without a teacher) and results of the automatic construction of the theory of recognition algorithms are presented. A description is given of the fundamental modules of the package of applied programs for solving recognition and classification problems (PARK), developed at the Computing Centre of the Academy of Sciences of the USSR on the basis of the algorithms described.