Universal method of accounting the properties existence constraints it problems of deriving formal concepts from fuzzy object-attributive data

Object of research is the processing of fuzzy object-attributive data — empirically established “object-properties” relation — in order to identify formal concepts and construct a lattice that fixes a partial ordered generalization relation on a set of formal concepts. The paper is devoted to implementation of the method of preprocessing the initial fuzzy “object-properties” relation proposed earlier by the authors, which provides for taking into account the properties existence constraints - a priori subjective attitudes that fix the permissible variants of the intent of formal concepts. The method is suitable for all types of biclustering tasks: to derive distinct and fuzzy concepts from the threshold cross-section of the source data, to derive fuzzy concepts based on the fuzzy closure of the source data. Implementation of the method involves the representation of a fuzzy set of properties for each object in the training sample as a composition of a finite kit of distinct sets, filtering this kit of sets for accounting properties existence constraints and approximation of a fuzzy set of properties of the object of training sample based on the composition of the filtered sets of properties. The effectiveness of the proposed implementation provides by the rational organization of each of these actions and in fact their partial combination.