TY - JOUR

T1 - Least-squares bilinear clustering of three-way data

AU - Schoonees, Pieter C.

AU - Groenen, Patrick J.F.

AU - van de Velden, Michel

N1 - Publisher Copyright:
© 2021, The Author(s).

PY - 2021/11/15

Y1 - 2021/11/15

N2 - 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.

AB - 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.

UR - http://www.scopus.com/inward/record.url?scp=85119049194&partnerID=8YFLogxK

U2 - 10.1007/s11634-021-00475-2

DO - 10.1007/s11634-021-00475-2

M3 - Article

AN - SCOPUS:85119049194

SN - 1862-5347

SP - 1001

EP - 1037

JO - Advances in Data Analysis and Classification

JF - Advances in Data Analysis and Classification

ER -