Classification of electroencephalogram records related to cursor movements with a hybrid method based on deep learning

Abstract

In brain computer interface (BCI), many transformation methods are used when processing electroencephalogram (EEG) signals. Thus, the EEG can be represented in different domains. However, designing an EEG‐based BCI system without any transformation technique is a challenge. For this purpose, in this study, a BCI model is proposed without any transformation. The classification of cursor down and cursor up movements using the EEG signals received from the brain is aimed at in the proposed model. The EEG patterns were classified using two methods. Firstly, EEG signals were classified by classic convolutional neural network (CNN). Secondly, proposed hybrid structure obtained the EEG features, which were classified by k‐NN and SVM, using CNN. Classification with CNN architecture gave a result of 68.15% while the hybrid method using k‐NN and SVM classifiers yielded 97.55% and 97.61% respectively. The hybrid proposed method were more successful than the studies in the literature.