Maruf, LafizAlam Z.Rahman M.M.E.N.Rahman, Md AnisurHough, PeterParvezz M.Z.2026-08-112026-08-112020-12-16L. Maruf, Z. Alam, M. M. -E. -N. Rahman, M. A. Rahman, P. Hough and M. Z. Parvezz, "Classification of Motor Imagery Tasks for Brain-Computer Interface using EEG Signals," 2020 IEEE Asia-Pacific Conference on Computer Science and Data Engineering (CSDE), Gold Coast, Australia, 2020, pp. 1-5, doi: 10.1109/CSDE50874.2020.9411545.97816654197412-s2.0-85105479718https://hdl.handle.net/10361/28908The human brain is the most important and central organ of the body. The brain receives information from the environment through the sensory organs, processes, analyses and integrates the information and sends instructions to the rest of the organs. There is also communication between billions of neurons within the brain for emotion, thoughts and behaviour. Brain-computer interface (BCI) is a communication medium that translates neuronal activity into commands towards controlling of an external system. Electroencephalogram (EEG) records the electrical activity of the brain by evaluating voltage changes in regions of the skin by simply placing the electrodes on the skin. As there are many disabled people for whom this process may help by activating movement in their limbs. Thus, we decided to work on classifying the motor imagery tasks using EEG signals. This research presented the process of classifying three motor imagery tasks using EEG signals which can be further evolved into a BCI system that can remotely control external devices. Different bands are filtered from EEG signals in order to extract different frequency distributed features for seven subjects who participated in this experiment. Two sets of features are used to classify different motor imagery tasks based on Support Vector Machine (SVM), Artificial Neural Networks (ANN), Decision Tree, Logistic Regression and Naive Bayes. The experimental results show that SVM achieved higher accuracy compared to ANN. Decision Tree, Logistic Regression, and Naive Bayes classifiers. The accuracy of our proposed method is also shown to be better than two other existing Motor Imagery classification techniques.5 Pagesen-USSupport vector machinesArtificial neural networksFeature extractionBrain modelingElectroencephalographyBrain-computer interfaces.Electroencephalography.Neural networks (Computer science).Conference Proceeding10.1109/CSDE50874.2020.9411545