A Rule Extraction Technique Applied to Ensembles of Neural Networks, Random Forests, and Gradient-Boosted Trees

In machine learning, ensembles of models based on Multi-Layer Perceptrons (MLPs) or decision trees are considered successful models. However, explaining their responses is a complex problem that requires the creation new methods interpretation. A natural way to explain classifications transform them into propositional rules. this work, we focus random forests and gradient-boosted trees. Specifically, these converted an ensemble interpretable MLPs from which rules produced. The rule extraction method presented here allows one precisely locate discriminating hyperplanes constitute antecedents experiments eight classification problems, compared our technique “Skope-Rules” other state-of-the-art techniques. Experiments were performed with ten-fold cross-validation trials, also generated MLPs. By evaluating characteristics extracted in terms complexity, fidelity, accuracy, results obtained showed competitive. To best knowledge, few works showing has been applied both neural networks.