Leveraging the trade-off between accuracy and interpretability in a hybrid intelligent system

Abstract

Neural Fuzzy Inference System (NFIS) is a widely adopted paradigm to develop a data-driven learning system. This hybrid system has been widely adopted due to its accurate reasoning procedure and comprehensible inference rules. Although most NFISs primarily focus on accuracy, we have observed an ever increasing demand on improving the interpretability of NFISs and other types of machine learning systems. In this paper, we illustrate how we leverage the trade-off between accuracy and interpretability in an NFIS called Genetic Algorithm and Rough Set Incorporated Neural Fuzzy Inference System (GARSINFIS). In a nutshell, GARSINFIS self-organizes its network structure with a small set of control parameters and constraints. Moreover, its autonomously generated inference rule base tries to achieve higher interpretability without sacrificing accuracy. Furthermore, we demonstrate different configuration options of GARSINFIS using well-known benchmarking datasets. The performance of GARSINFIS on both accuracy and interpretability is shown to be encouraging when compared against other decision tree, Bayesian, neural and neural fuzzy models.