CADM: Cluster-customized Adaptive Distance Metric for Categorical Data Clustering

TL;DR

This paper introduces CADM, a novel cluster-customized distance metric for categorical and mixed data clustering that adapts to attribute distributions within each cluster, improving clustering accuracy.

Contribution

The paper proposes a new adaptive distance metric tailored for categorical and mixed data, accounting for cluster-specific attribute distributions, which enhances clustering performance.

Findings

Achieves top ranking in 14 datasets

Effectively handles mixed numerical and categorical data

Demonstrates improved clustering accuracy

Abstract

An appropriate distance metric is crucial for categorical data clustering, as the distance between categorical data cannot be directly calculated. However, the distances between attribute values usually vary in different clusters induced by their different distributions, which has not been taken into account, thus leading to unreasonable distance measurement. Therefore, we propose a cluster-customized distance metric for categorical data clustering, which can competitively update distances based on different distributions of attributes in each cluster. In addition, we extend the proposed distance metric to the mixed data that contains both numerical and categorical attributes. Experiments demonstrate the efficacy of the proposed method, i.e., achieving an average ranking of around first in fourteen datasets. The source code is available at https://anonymous.4open.science/r/CADM-47D8