Making sense of multiple distance matrices through common and distinct components

Multiblock analysis attacks the problem of how to combine data from various sources for purposes such as prediction, classification, clustering, or visual analysis. A key concept is distinction between “common” and “distinct” parts, that is, what information repeats itself across blocks unique an individual block. The statistical field multiblock holds many different approaches, which leads treatments both terms distinct common themselves differences in numerical results. In this article, we extend discussion domain distance matrices, situation where point sets, so-called configurations, are analyzed via relative distances either because configurations not available directly a representation favorable. Situations typical chemometrics will be highlighted illustrated examples. When analyzing methods, have focused on three aspects. First, during transition configuration domains, one needs consider multiple matrices treated. Second, when extracting manage tradeoff explaining variance ensuring similarity subspaces. Third, there design choice made whether subspace containing parts “shared” if separate subspaces associated with each aspects help categorize explain well-known methods field. selection was subsequently applied