Comparative Analysis for the Influence of the Tuning Parameters in the Algorithm for Detection of Epilepsy Based on Fuzzy Neural Networks

Abstract

This study presents a comparative analysis for the influence of the tuning parameters in our previously published algorithm for detection of epilepsy [2]. As the algorithm in [2] is generated using wavelet transform (WT) for feature extraction, and Adaptive Neuro-Fuzzy Inference System (ANFIS) for classification, the comparison in this paper is based on the different data splitting methods, the different input space partitioning methods in the ANFIS model, the usage of the different wavelet functions in the WT, the effects of normalization, as well as the effects of using different membership functions. The model was evaluated in terms of training performance and classification accuracies, and it was concluded that different combinations of input parameters differently classify the EEG signals.