DOI: https://doi.org/10.15587/1729-4061.2014.22930

Some methods of automatic grouping of objects

Наталія Емерихівна Кондрук

Abstract


Cluster analysis is relevant and widely used in information systems, medicine, psychology, chemistry, biology, public administration, philology, marketing, sociology and other disciplines. However, the wide use causes coherence and unambiguity problems of the mathematical apparatus for cluster analysis. In particular, taking into account that clustering data can have different physical meaning and that the objects similarity criteria are not universal and can be defined for different applied problems in different ways, building alternative (to the already known) similarity coefficients, which meet the emerging needs for grouping objects of new applied problems is relevant. Therefore, the purpose of the paper is to improve the efficiency of solving the cluster analysis problems by developing general methods and algorithms for clustering objects based on the "angular" and ''length" metrics and binary relations. General method for clustering objects based on fuzzy binary relations is developed in the paper. Semimetrics, characterizing the proximity degree of vectors of object features by the "angular" and "length" similarity are determined. Clustering algorithms, based on grouping objects by the introduced angular and length semimetrics are built. Software implementation of this method has shown its effectiveness in solving various applied problems and ease of use.


Keywords


cluster analysis; cluster; fuzzy binary relations; objects splitting; clustering objects

References


Estivill-Castro, V. Why so many clustering algorithms — A Position Paper [Text] / V. Estivill-Castro // ACM SIGKDD Explorations Newsletter. – 2002. – Vol. 4 (1). – P. 65–75.

Huang, Z. Extensions to the k-means algorithm for clustering large data sets with categorical values [Text] / Z. Huang // Data Mining and Knowledge Discovery. – 1998. – Vol. 2. – P. 283–304.

Mingoti, S. Comparing SOM neural network with Fuzzy c-means, K-means and traditional hierarchical clustering algorithms [Text] / S. Mingoti, J. Lima // European Journal of Operational Research. – 2006. – Vol. 174 (3). – P. 1742–1759.

Székely, G. J. Hierarchical clustering via Joint Between-Within Distances: Extending Ward’s Minimum Variance Method [Text] / G. J. Székely, M. L. Rizzo // Journal of Classification. – 2005. – Vol. 22. – P. 151–183.

Bailey, K. Numerical Taxonomy and Cluster Analysis [Text] / K. Bailey. – Typologies and Taxonomies, 1994. – 34 p.

Jain, A. K. Flynn Data clustering: a review [Text] / A. K. Jain, M. N. Murty // ACM Comput. Surv. – 1999. – Vol. 31(3). – P. 264–323.

Пістунов, І. М. Кластерний аналіз в економіці [Текст] / І. М. Пістунов, О. П. Антонюк та ін. – Дніпропетровськ: Національний гірничий університет, 2008.– 84 с.

Ким, Дж. Факторный, дискриминантный и кластерний анализ [Текст] / Дж. Ким, Ч. У. Мьюллер, У. Р. Клекка. – М.: Финансы и статистика, 1989. – 215 с.

Дюран, Б. Кластерный анализ [Текст] / Б. Дюран, П. Оделл. – М.: «Статистика», 1977. – 128 с.

Кондрук, Н. Е. Застосування багатокритеріальних моделей для задач збалансованого харчування [Текст] / Н. Е. Кондрук, М. М. Маляр // Вісник Черкаського державного технологічного університету. Серія: технічні науки. – 2010. – Вип. 1, № 1. – С. 3–7.

Кондрук, Н. Э. Некоторые применения кластеризации критериального пространства для задач выбора [Текст] / Н. Э. Кондрук, Н. Н. Маляр // Компьютерная математика. – 2009. – № 2. – С. 142–149.

А61К8/19, А61К8/30, МПК (2006.01). Патент на корисну модель 64777 Україна. Спосіб автоматизованого складання дієтичного харчування «Дієтолог» [Текст] / Маляр М. М., Кондрук Н. Е., Горленко О. М., Томей А.І . – № u201100007; Заявл. від 04.01.2011; Опубл. 25.11.2011, Бюл.№ 22.

Estivill-Castro, V. (2002). Why so many clustering algorithms — A Position Paper. ACM SIGKDD Explorations Newsletter, 4 (1), 65–75.

Huang, Z. (1998). Extensions to the k-means algorithm for clustering large data sets with categorical values. Data Mining and Knowledge Discovery, 2, 283–304.

Mingoti, S., Lima, J. (2006). Comparing SOM neural network with Fuzzy c-means, K-means and traditional hierarchical clustering algorithms. European Journal of Operational Research, 174 (3), 1742–1759.

Székely, G. J., Rizzo, M. L. (2005). Hierarchical clustering via Joint Between-Within Distances: Extending Ward’s Minimum Variance Method. Journal of Classification, 22, 151–183.

Bailey, Ken (1994). Numerical Taxonomy and Cluster Analysis. Typologies and Taxonomies, 34.

Jain, A. K., Murty, M. N. (1999). Flynn Data clustering: a review. ACM Comput. Surv., 31 (3), 264–323.

Pistunov, I. M. (2008). Cluster analysis of the economy. National Mining University, 84.

Durand, B. (1977). Cluster analysis. “Statistics”, 128.

Kim, J. (1989). Factor, discriminant and cluster analysis. Finance and Statistics, 215.

Kondruk, N. E. (2010). Application of multicriteria models for the problems of a balanced diet. J of Cherkasy State Technological University. Series: Engineering Sciences, Vol. 1, № 1, 3–7.

Kondruk, N. E. (2009). Some applications of clustering criterion space for selection tasks. Computer Mathematics, 2, 142–149.

Malyar, M. M., Kondruk, N. E., Gorlenko, A. M., Tomey A. A. (25.11.2011). Ukraine Automated method dietetic foods “Nutritionist”. Patent for utility model 64777 u201100007, № 22.


GOST Style Citations








Copyright (c) 2014 Наталія Емерихівна Кондрук

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

ISSN (print) 1729-3774, ISSN (on-line) 1729-4061