期刊
ADVANCES IN DATA ANALYSIS AND CLASSIFICATION
卷 16, 期 4, 页码 1001-1037出版社
SPRINGER HEIDELBERG
DOI: 10.1007/s11634-021-00475-2
关键词
Three-way data; Bilinear decomposition; k-Means cluster analysis; Least-squares estimation; Biplots
A least-squares bilinear clustering framework is introduced to model three-way data, combining bilinear decompositions of two-way matrices with clustering over observations. Different clusterings are defined for each part of the bilinear decomposition, reducing computational burden. The method simplifies the joint clustering problem into separate problems that can be handled independently.
A least-squares bilinear clustering framework for modelling three-way data, where each observation consists of an ordinary two-way matrix, is introduced. The method combines bilinear decompositions of the two-way matrices with clustering over observations. Different clusterings are defined for each part of the bilinear decomposition, which decomposes the matrix-valued observations into overall means, row margins, column margins and row-column interactions. Therefore up to four different classifications are defined jointly, one for each type of effect. The computational burden is greatly reduced by the orthogonality of the bilinear model, such that the joint clustering problem reduces to separate problems which can be handled independently. Three of these sub-problems are specific cases of k-means clustering; a special algorithm is formulated for the row-column interactions, which are displayed in clusterwise biplots. The method is illustrated via an empirical example and interpreting the interaction biplots are discussed. Supplemental materials for this paper are available online, which includes the dedicated R package, lsbclust.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据