Neurosymbolic Association Rule Mining from Tabular Data

2025-04-27

AI Summary

Mining association rules (ARM) in high-dimensional data often triggers rule explosion, leading to poor computational efficiency and limited interpretability. To address this, we propose Aerial+, a neuro-symbolic integration framework that employs an undercomplete autoencoder to learn feature-correlation representations and automatically distills high-quality logical rules via reconstruction-driven optimization—introducing the novel paradigm of “neural-representation-guided rule generation.” Aerial+ simultaneously ensures conciseness, completeness, and quality of the rule set while overcoming the scalability limitations of traditional symbolic ARM methods in high-dimensional settings. Evaluated on five benchmark datasets, Aerial+ consistently outperforms seven state-of-the-art baselines: it reduces rule count substantially while achieving 100% data coverage. When embedded into downstream models, it accelerates inference without sacrificing accuracy—maintaining or even improving classification performance.

Technology Category

• Data Mining & Knowledge Management: Rule Mining & Pattern Mining
• Machine Learning: Neuro-Symbolic Learning
• Knowledge Representation and Reasoning: Automated Reasoning and Theorem Proving

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Association Rule Mining (ARM) is the task of mining patterns among data features in the form of logical rules, with applications across a myriad of domains. However, high-dimensional datasets often result in an excessive number of rules, increasing execution time and negatively impacting downstream task performance. Managing this rule explosion remains a central challenge in ARM research. To address this, we introduce Aerial+, a novel neurosymbolic ARM method. Aerial+ leverages an under-complete autoencoder to create a neural representation of the data, capturing associations between features. It extracts rules from this neural representation by exploiting the model's reconstruction mechanism. Extensive evaluations on five datasets against seven baselines demonstrate that Aerial+ achieves state-of-the-art results by learning more concise, high-quality rule sets with full data coverage. When integrated into rule-based interpretable machine learning models, Aerial+ significantly reduces execution time while maintaining or improving accuracy.

Problem

• Addresses rule explosion in high-dimensional Association Rule Mining
• Improves execution time and accuracy of rule-based models
• Develops concise, high-quality rule sets with full data coverage

Innovation

• Uses under-complete autoencoder for neural representation
• Extracts rules via model reconstruction mechanism
• Produces concise high-quality rule sets efficiently