Categorical Data Clustering via Value Order Estimated Distance Metric Learning

TL;DR

This paper introduces a novel order distance metric learning approach for clustering categorical data, which improves accuracy and interpretability by learning optimal value order relationships and integrating them into a unified clustering framework.

Contribution

It proposes a new joint learning paradigm that alternates between clustering and order distance metric learning, specifically designed for categorical and mixed datasets.

Findings

Achieves superior clustering accuracy on categorical datasets

Reduces complexity in understanding categorical data

Validated through experiments, ablation studies, and significance tests

Abstract

Clustering is a popular machine learning technique for data mining that can process and analyze datasets to automatically reveal sample distribution patterns. Since the ubiquitous categorical data naturally lack a well-defined metric space such as the Euclidean distance space of numerical data, the distribution of categorical data is usually under-represented, and thus valuable information can be easily twisted in clustering. This paper, therefore, introduces a novel order distance metric learning approach to intuitively represent categorical attribute values by learning their optimal order relationship and quantifying their distance in a line similar to that of the numerical attributes. Since subjectively created qualitative categorical values involve ambiguity and fuzziness, the order distance metric is learned in the context of clustering. Accordingly, a new joint learning paradigm is developed to alternatively perform clustering and order distance metric learning with low time complexity and a guarantee of convergence. Due to the clustering-friendly order learning mechanism and the homogeneous ordinal nature of the order distance and Euclidean distance, the proposed method achieves superior clustering accuracy on categorical and mixed datasets. More importantly, the learned order distance metric greatly reduces the difficulty of understanding and managing the non-intuitive categorical data. Experiments with ablation studies, significance tests, case studies, etc., have validated the efficacy of the proposed method. The source code is available at https://github.com/DAJ0612/OCL_Source_Code.