Effectiveness of Deep Image Embedding Clustering Methods on Tabular Data

TL;DR

This study evaluates deep embedding clustering methods on tabular data, revealing traditional clustering often outperforms deep learning approaches, emphasizing the need for data-centric customization.

Contribution

First comprehensive assessment of deep embedding clustering methods on tabular data, highlighting their limitations and the robustness of traditional clustering techniques.

Findings

Traditional clustering ranks second among tested methods.

Deep embedding clustering methods are often outperformed by conventional approaches.

Data-centric customization is essential for deep methods to be competitive.

Abstract

Deep learning methods in the literature are commonly benchmarked on image data sets, which may not be suitable or effective baselines for non-image tabular data. In this paper, we take a data-centric view to perform one of the first studies on deep embedding clustering of tabular data. Eight clustering and state-of-the-art embedding clustering methods proposed for image data sets are tested on seven tabular data sets. Our results reveal that a traditional clustering method ranks second out of eight methods and is superior to most deep embedding clustering baselines. Our observation aligns with the literature that conventional machine learning of tabular data is still a robust approach against deep learning. Therefore, state-of-the-art embedding clustering methods should consider data-centric customization of learning architectures to become competitive baselines for tabular data.