![[2602.13586] Interpretable clustering via optimal multiway-split decision trees](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.13586] Interpretable clustering via optimal multiway-split decision trees
Summary
This paper presents a novel clustering method using optimal multiway-split decision trees, enhancing interpretability and accuracy while reducing computational costs compared to traditional binary decision trees.
Why It Matters
Clustering is essential for understanding complex data structures. This research addresses the limitations of existing methods, offering a more interpretable and efficient approach that could significantly benefit fields relying on data analysis, such as machine learning and data science.
Key Takeaways
- Introduces optimal multiway-split decision trees for clustering.
- Improves interpretability and accuracy over traditional binary trees.
- Utilizes a one-dimensional K-means algorithm for effective variable discretization.
- Demonstrates superior performance through extensive numerical experiments.
- Addresses computational inefficiencies in existing clustering methods.
Computer Science > Machine Learning arXiv:2602.13586 (cs) [Submitted on 14 Feb 2026] Title:Interpretable clustering via optimal multiway-split decision trees Authors:Hayato Suzuki, Shunnosuke Ikeda, Yuichi Takano View a PDF of the paper titled Interpretable clustering via optimal multiway-split decision trees, by Hayato Suzuki and 1 other authors View PDF HTML (experimental) Abstract:Clustering serves as a vital tool for uncovering latent data structures, and achieving both high accuracy and interpretability is essential. To this end, existing methods typically construct binary decision trees by solving mixed-integer nonlinear optimization problems, often leading to significant computational costs and suboptimal solutions. Furthermore, binary decision trees frequently result in excessively deep structures, which makes them difficult to interpret. To mitigate these issues, we propose an interpretable clustering method based on optimal multiway-split decision trees, formulated as a 0-1 integer linear optimization problem. This reformulation renders the optimization problem more tractable compared to existing models. A key feature of our method is the integration of a one-dimensional K-means algorithm for the discretization of continuous variables, allowing for flexible and data-driven branching. Extensive numerical experiments on publicly available real-world datasets demonstrate that our method outperforms baseline methods in terms of clustering accuracy and interpretabilit...
