The Use of Fuzzy Inference Systems for Classification in EEG-based Brain-Computer Interfaces

This paper introduces the use of a Fuzzy Inference System (FIS) for classification in EEG-based Brain-Computer Interfaces (BCI) systems. We present our FIS algorithm and compare it, on motor imagery signals, with three other popular classifiers, widely used in the BCI community. Our results show that FIS outperformed a Linear Classifier and reached the same level of accuracy as Support Vector Machine and neural networks. Thus, FIS-based classification is suitable for BCI design. Furthermore, FIS algorithms have two additionnal advantages: they are readable and easily extensible.