Optimal Decision Trees for Interpretable and Constrained Clustering
Bibliographic record
Abstract
Constrained clustering is a semi-supervised approach to determining meaningful groupings of data that respect userspecified constraints. Such constraints are typically used to enforce desirable structural and domain-specific properties of the resulting clusters. Notably, such constraints can significantly improve the quality and accuracy of clustering. Data clustering solutions can take on many different forms. Decision trees are a particularly desirable solution form because of their inherent interpretability. Unfortunately, existing decision tree clustering approaches do not support clustering constraints and do not provide strong theoretical guarantees with respect to solution quality. To address the task of decision tree clustering with constraints, we present a novel SAT-based encoding that solves the problem to an approximated optimality in relation to a well-known bi-criteria objective. Our framework is the first exact approach for interpretable constrained clustering with decision trees. Experiments involving a range of real-world and synthetic datasets demonstrate that our approach can produce interpretable clustering solutions that are of superior quality compared to their non-interpretable counterparts, with or without the addition of constraints. We further provide new insights into the trade-off between interpretability and the satisfaction of user-specified constraints, presenting extensions to our clustering approach that treat the satisfaction of constraints as an additional optimization objective.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.017 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.002 | 0.003 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.002 | 0.002 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.005 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".