Article

Article title PROBLEM OF THE CHOICE OF CRITERIA IN METHODS OF THE FUZZY CLUSTERING
Authors A.V. Egorov, N.I. Kupriyanova
Section SECTION VI. NEW INFORMATION TECHNOLOGIES IN POWER
Month, Year 02, 2013 @en
Index UDC 004.4
DOI
Abstract Research objective is the description of the problem caused by variety of existing criteria in the indistinct cluster analysis. The subgroups of criteria serving for definition of clusters, and also for an assessment of process of a clustering and its results are described. The main criteria are illustrated on basic types of data: matrixes of communication and matrixes object – a sign. Further the mechanism of calculation of these criteria is considered. In end these criteria correspond to criterion of k-averages, as one of basic criteria of the indistinct cluster analysis.

Download PDF

Keywords Optimizing criteria; criteria of a choice of methods of a clustering; functionalities of quality of a clustering.
References 1. Егоров А.В., Куприянова Н.И. Особенности методов кластеризации данных // Известия ЮФУ. Технические науки. – 2011. – № 11 (124). – С. 174-178.
2. Миркин Б.Г. Методы кластер – анализа для поддержки принятия решений: обзор. – М.: Высшая школа экономики, 2011.

Comments are closed.