Abstract
Brain-Computer Interface (BCI) is a new technology that uses electrodes and sensors to connect machines and computers with the human brain to improve a person's mental performance. Also, human intentions and thoughts are analyzed and recognized using BCI, which is then translated into Electroencephalogram (EEG) signals. However, certain brain signals may contain redundant information, making classification ineffective. Therefore, relevant characteristics are essential for enhancing classification performance.. Thus, feature selection has been employed to eliminate redundant data before sorting to reduce computation time. BCI Competition III Dataset Iva was used to investigate the efficacy of the proposed system. A Smart Bagged Tree-based Classifier (SBT-RF) technique is presented to determine the importance of the features for selecting and classifying the data. As a result, SBT-RF is better at improving the mean accuracy of the dataset. It also decreases computation cost and training time and increases prediction speed. Furthermore, fewer features mean fewer electrodes, thus lowering the risk of damage to the brain. The proposed algorithm has the greatest average accuracy of similar to 98% compared to other relevant algorithms in the literature. SBT-RF is compared to state-of-the-art algorithms based on the following performance metrics: Confusion Matrix, ROC-AUC, F1-Score, Training Time, Prediction speed, and Accuracy.