Enhancing interpretability of rule-based classifiers through feature graphs

TL;DR

This paper introduces a new framework that enhances the interpretability of complex rule-based classifiers by visualizing feature interactions, providing a novel importance metric, and enabling comparison of rule sets, with applications in clinical data analysis.

Contribution

It presents a comprehensive, graph-based approach for estimating feature contributions and comparing rule-based models, improving interpretability and insight discovery in sensitive domains.

Findings

Effective in uncovering clinical feature interactions

Demonstrates superior robustness over existing methods

Applicable across multiple rule-based algorithms

Abstract

In domains where transparency and trustworthiness are crucial, such as healthcare, rule-based systems are widely used and often preferred over black-box models for decision support systems due to their inherent interpretability. However, as rule-based models grow complex, discerning crucial features, understanding their interactions, and comparing feature contributions across different rule sets becomes challenging. To address this, we propose a comprehensive framework for estimating feature contributions in rule-based systems, introducing a graph-based feature visualisation strategy, a novel feature importance metric agnostic to rule-based predictors, and a distance metric for comparing rule sets based on feature contributions. By experimenting on two clinical datasets and four rule-based methods (decision trees, logic learning machines, association rules, and neural networks with rule extraction), we showcase our method's capability to uncover novel insights on the combined predictive value of clinical features, both at the dataset and class-specific levels. These insights can aid in identifying new risk factors, signature genes, and potential biomarkers, and determining the subset of patient information that should be prioritised to enhance diagnostic accuracy. Comparative analysis of the proposed feature importance score with state-of-the-art methods on 15 public benchmarks demonstrates competitive performance and superior robustness. The method implementation is available on GitHub: https://github.com/ChristelSirocchi/rule-graph.