Clustering categorical data via ensembling dissimilarity matrices

TL;DR

This paper introduces a novel clustering method for categorical data that uses ensembling of dissimilarity matrices, demonstrating its effectiveness across various data dimensions and types, including genome sequences.

Contribution

The paper proposes a new ensembling approach for clustering categorical data, extending to high dimensions and unequal lengths, with demonstrated advantages over existing methods.

Findings

Effective on low-dimensional categorical data

Performs well on high-dimensional data with equal lengths

Outperforms phylogenetic tree clusterings on genome sequences

Abstract

We present a technique for clustering categorical data by generating many dissimilarity matrices and averaging over them. We begin by demonstrating our technique on low dimensional categorical data and comparing it to several other techniques that have been proposed. Then we give conditions under which our method should yield good results in general. Our method extends to high dimensional categorical data of equal lengths by ensembling over many choices of explanatory variables. In this context we compare our method with two other methods. Finally, we extend our method to high dimensional categorical data vectors of unequal length by using alignment techniques to equalize the lengths. We give examples to show that our method continues to provide good results, in particular, better in the context of genome sequences than clusterings suggested by phylogenetic trees.