Classification of EEG Signals using Various Machine-Learning Techniques and Essential Outcomes

Electroencephalography (EEG) signal classification plays a pivotal role in various applications, including medical diagnostic, brain-computer interface (BCI), and cognitive state monitoring. A comparative analysis of three prominent Machine Learning (ML) techniques was carried out, namely Convolutional Neural Networks (CNN), Support Vector Machines(SVM),and Linear Discriminant Analysis (LDA) for EEG signals classification. Signal categorization has drawn more and more attention as a crucial problem in BCI systems. The present research introduces a novel deep CNN based motor imagery MI-EEG classification technique. The findings show that CNN can perform better in classification when compared to the other two techniques (SVM and LDA). The overall accuracy using CNN is 96.05 ± 1.192 %, which is higher than SVM (93.64 ± 0.7871 %), and LDA (92.72 ± 0.7366 %). According to the current research, the suggested approach for classifying MI is successful and may be a good fit for BCI applications.