Discovering Association Rules in High-Dimensional Small Tabular Data

2025-09-24

AI Summary

Association rule mining in high-dimensional, low-sample tabular data (e.g., gene expression data with ~18k features and 50 samples) suffers from combinatorial rule explosion, prohibitive computational cost, and severe performance degradation of existing methods (e.g., Aerial+). Method: This work formally defines association rule mining in this challenging regime and proposes two fine-tuning strategies leveraging tabular foundation models, integrated within the Aerial+ neuro-symbolic framework to enhance feature representation and rule generation. Contribution/Results: Experiments on five real-world datasets demonstrate substantial improvements in rule quality and interpretability. The method achieves 1–2 orders of magnitude better scalability than state-of-the-art approaches, enabling effective knowledge discovery and interpretable machine learning in high-dimensional, data-scarce settings.

Technology Category

• Data Mining & Knowledge Management: Rule Mining & Pattern Mining
• Machine Learning: Neuro-Symbolic Learning
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Large pretrained models with web data
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Association Rule Mining (ARM) aims to discover patterns between features in datasets in the form of propositional rules, supporting both knowledge discovery and interpretable machine learning in high-stakes decision-making. However, in high-dimensional settings, rule explosion and computational overhead render popular algorithmic approaches impractical without effective search space reduction, challenges that propagate to downstream tasks. Neurosymbolic methods, such as Aerial+, have recently been proposed to address the rule explosion in ARM. While they tackle the high dimensionality of the data, they also inherit limitations of neural networks, particularly reduced performance in low-data regimes. This paper makes three key contributions to association rule discovery in high-dimensional tabular data. First, we empirically show that Aerial+ scales one to two orders of magnitude better than state-of-the-art algorithmic and neurosymbolic baselines across five real-world datasets. Second, we introduce the novel problem of ARM in high-dimensional, low-data settings, such as gene expression data from the biomedicine domain with around 18k features and 50 samples. Third, we propose two fine-tuning approaches to Aerial+ using tabular foundation models. Our proposed approaches are shown to significantly improve rule quality on five real-world datasets, demonstrating their effectiveness in low-data, high-dimensional scenarios.

Problem

• Addressing rule explosion and computational overhead in high-dimensional association rule mining
• Solving association rule mining challenges in high-dimensional, low-data tabular settings
• Improving rule quality for small tabular data with many features using fine-tuning

Innovation

• Scales significantly better than existing baselines
• Introduces ARM for high-dimensional low-data settings
• Proposes fine-tuning approaches using tabular foundation models