Overview

PyRuleAnalyzer is a Python-based tool designed to support rule extraction, analysis, and simplification from decision tree and random forest models. It provides a comprehensive pipeline to generate interpretable models, remove redundancies, and evaluate model accuracy and interpretability.

Key functionalities include:

• Extracting decision rules from Random Forests and Decision Trees.

• Identifying and removing redundant and duplicate rules.

• Evaluating initial and simplified models through classification accuracy, confusion matrices, and interpretability metrics.

• Computing sparsity and interpretability scores to assess model complexity and understandability.

• Saving, loading, and applying rule-based classifiers on new datasets.