Open AccessRobust method for finding the center and the scatter matrix of the cluster
Author(s) -
З. М. Шибзухов
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1479/1/012045
Subject(s) - mahalanobis distance , outlier , robustness (evolution) , differentiable function , mathematics , scatter matrix , distance matrix , algorithm , matrix (chemical analysis) , robust statistics , computer science , statistics , covariance matrix , mathematical analysis , biochemistry , chemistry , materials science , estimation of covariance matrices , composite material , gene
The problem of finding the centers and scattering matrices for a finite set of points containing outliers in a multidimensional space is considered. A new approach is considered in which instead of the arithmetic mean, differentiable mean values are used that are insensitive to outliers. An iterative reweighting scheme for searching for centers and corresponding scattering matrices for the Mahalanobis distance is considered. The examples presented in the article show the robustness property of the proposed method and algorithm with respect to a large number of outliers.