# Weight-Informed Self-Explaining Clustering for Mixed-Type Tabular Data

Source: [Weight-Informed Self-Explaining Clustering for Mixed-Type Tabular Data](<https://www.sota2.com/research/paper/69d5e1397d0172140ec75c4d>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-09T05:00:13.067Z

Authors: Lehao Li, Qiang Huang, Yihao Ang, Bryan Kian Hsiang Low, Anthony K. H. Tung, Xiaokui Xiao

Published: 2026-04-07T13:18:31.000Z

## Abstract

Clustering mixed-type tabular data is fundamental for exploratory analysis, yet remains challenging due to misaligned numerical-categorical representations, uneven and context-dependent feature relevance, and disconnected and post-hoc explanation from the clustering process. We propose WISE, a Weight-Informed Self-Explaining framework that unifies representation, feature weighting, clustering, and interpretation in a fully unsupervised and transparent pipeline. WISE introduces Binary Encoding with Padding (BEP) to align heterogeneous features in a unified sparse space, a Leave-One-Feature-Out (LOFO) strategy to sense multiple high-quality and diverse feature-weighting views, and a two-stage weight-aware clustering procedure to aggregate alternative semantic partitions. To ensure intrinsic interpretability, we further develop Discriminative FreqItems (DFI), which yields feature-level explanations that are consistent from instances to clusters with an additive decomposition guarantee. Extensive experiments on six real-world datasets demonstrate that WISE consistently outperforms classical and neural baselines in clustering quality while remaining efficient, and produces faithful, human-interpretable explanations grounded in the same primitives that drive clustering.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Mixed-type Tabular Clustering on Adult](<https://www.sota2.com/research/sota/mixed-type-tabular-clustering-on-adult.md>): Mixed-type tabular clustering; Adult; ARI: 0.663.

- [Mixed-type tabular clustering on Vermont](<https://www.sota2.com/research/sota/mixed-type-tabular-clustering-on-vermont.md>): Mixed-type tabular clustering; Vermont; ARI: 0.283.

- [Mixed-type tabular clustering on Arizona](<https://www.sota2.com/research/sota/mixed-type-tabular-clustering-on-arizona.md>): Mixed-type tabular clustering; Arizona; ARI: 0.309.

- [Mixed-type Tabular Clustering on Obesity](<https://www.sota2.com/research/sota/mixed-type-tabular-clustering-on-obesity.md>): Mixed-type tabular clustering; Obesity; ARI: 0.222.

- [Mixed-type tabular clustering on Credit](<https://www.sota2.com/research/sota/mixed-type-tabular-clustering-on-credit.md>): Mixed-type tabular clustering; Credit; ARI: 0.328.

- [Mixed-type tabular clustering on GeoNames](<https://www.sota2.com/research/sota/mixed-type-tabular-clustering-on-geonames.md>): Mixed-type tabular clustering; GeoNames; ARI: 0.146.

- [Clustering on Adult, Vermont, Arizona, Obesity, Credit, and GeoNames Average Rank (test)](<https://www.sota2.com/research/sota/clustering-on-adult-vermont-arizona-obesity-credit-and-geonames-average-rank-test.md>): Clustering; Adult, Vermont, Arizona, Obesity, Credit, and GeoNames Average Rank (test); ARI: 1.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.05857>)

[PDF](<https://arxiv.org/pdf/2604.05857.pdf>)
