Performance-Interpretability Tradeoff of Mamdani Neuro-Fuzzy Classifiers for Medical Data

Neuro-fuzzy systems are models that incorporate the learning ability and performance of Artificial Neural Networks (ANNs) with the interpretable reasoning of fuzzy inference systems (FISs). An ANN can learn patterns from data and achieve high accuracy, while a FIS uses linguistic and interpretable rules to match inputs and outputs of the data. Two types of FISs are used the most in literature: Takagi-Sugeno-Kang (TSK) and Mamdani. The main focus of this paper is on the Mamdani neuro-fuzzy systems, notably the Hybrid Neuro-Fuzzy Inference System (HyFIS) and the Neuro-Fuzzy Classifier (NEFCLASS). It aims at evaluating and comparing the two classifiers over two medical datasets to study their performance-interpretability tradeoff. Results show that HyFIS is the best in terms of performance, while NEFCLASS is better in terms of interpretability. As for the performance-interpretability tradeoff, NEFCLASS has the best overall results; it achieves a good performance while being less complicated and more interpretable. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.