FRRI: a novel algorithm for fuzzy-rough rule induction
Henri Bollaert, Marko Palangetić, Chris Cornelis, Salvatore Greco, Roman Słowiński
TL;DR
FRRI addresses the need for interpretable classification by merging fuzzy rule induction with rough-set premises to create compact, accurate rulesets. It introduces a two-stage process: rule shortening, which generalizes rules from each training object while preserving class, and rule selection through integer programming to minimize the final rule set while preserving full coverage. Experiments on 18 KEEL numerical datasets show FRRI achieves higher average balanced accuracy and smaller rule sets than state-of-the-art baselines (QuickRules, MODLEM, FURIA, RIPPER), with particular strength on imbalanced data. The work highlights FRRI’s potential for transparent decision models and outlines avenues for regression, ordering strategies, approximate solvers, bagging ensembles, and hierarchical rule integration.
Abstract
Interpretability is the next frontier in machine learning research. In the search for white box models - as opposed to black box models, like random forests or neural networks - rule induction algorithms are a logical and promising option, since the rules can easily be understood by humans. Fuzzy and rough set theory have been successfully applied to this archetype, almost always separately. As both approaches to rule induction involve granular computing based on the concept of equivalence classes, it is natural to combine them. The QuickRules\cite{JensenCornelis2009} algorithm was a first attempt at using fuzzy rough set theory for rule induction. It is based on QuickReduct, a greedy algorithm for building decision reducts. QuickRules already showed an improvement over other rule induction methods. However, to evaluate the full potential of a fuzzy rough rule induction algorithm, one needs to start from the foundations. In this paper, we introduce a novel rule induction algorithm called Fuzzy Rough Rule Induction (FRRI). We provide background and explain the workings of our algorithm. Furthermore, we perform a computational experiment to evaluate the performance of our algorithm and compare it to other state-of-the-art rule induction approaches. We find that our algorithm is more accurate while creating small rulesets consisting of relatively short rules. We end the paper by outlining some directions for future work.