SpecVAT: Enhanced visual cluster analysis

Abstract

Given a pairwise dissimilarity matrix D of a set of objects, visual methods such as the VAT algorithm (for visual analysis of cluster tendency) representD as an image I(D̃) where the objects are reordered to highlight cluster structure as dark blocks along the diagonal of the image. A major limitation of such visual methods is their inability to highlight cluster structure in I(D̃) when D contains clusters with highly complex structure. In this paper, we address this limitation by proposing a Spectral VAT (SpecVAT) algorithm, where D is mapped to D- in an embedding space by spectral decomposition of the Laplacian matrix, and then reordered to D̃' using the VAT algorithm. We also propose a st..