RuleKit 2: Faster and simpler rule learning

2025-04-29

AI Summary

RuleKit suffers from low computational efficiency, poor usability, and limited task coverage (e.g., lack of support for regression and survival analysis). To address these limitations, we propose RuleKit 2—the first high-performance, interpretable rule learning framework supporting classification, regression, and survival analysis. Methodologically, it features algorithmic enhancements to AQ and CN2, multi-threaded parallelization, and low-level Cython acceleration, achieving up to 100× speedup on certain tasks. It introduces a dual-interface design: a native Python package fully compliant with the scikit-learn API and a lightweight Streamlit-based web GUI. Additionally, it enables end-to-end visual modeling and rule interpretation. The open-source implementation (GNU AGPLv3) has been integrated into multiple research and industrial pipelines, substantially improving the practicality, accessibility, and adoption of rule-based learning tools.

Technology Category

• Data Mining & Knowledge Management: Rule Mining & Pattern Mining
• Machine Learning: Classification and Regression
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs

Abstract

Rules offer an invaluable combination of predictive and descriptive capabilities. Our package for rule-based data analysis, RuleKit, has proven its effectiveness in classification, regression, and survival problems. Here we present its second version. New algorithms and optimized implementations of those previously included, significantly improved the computational performance of our suite, reducing the analysis time of some data sets by two orders of magnitude. The usability of RuleKit 2 is provided by two new components: Python package and browser application with a graphical user interface. The former complies with scikit-learn, the most popular data mining library for Python, allowing RuleKit 2 to be straightforwardly integrated into existing data analysis pipelines. RuleKit 2 is available at GitHub under GNU AGPL 3 license (https://github.com/adaa-polsl/RuleKit)

Problem

• Enhances rule learning for classification, regression, survival tasks
• Improves computational performance and reduces analysis time
• Expands usability with Python package and browser GUI

Innovation

• Optimized algorithms for faster rule learning
• Python package compatible with scikit-learn
• Browser application with graphical interface