Classification of Concentration and Rest by Power Spectral Analysis with Support Vector Machine Model

Recognition of mental state is necessary for cognitive and behavioral studies. Concentration is one of the mental states that have a considerable role in daily activity. The efficiency of work will be improved when we are maintained high concentration. Besides, concentration state recognition also provides a promising alternative way in clinical to detect attention deficit hyperactive disorder. Monitoring concentration state has many methods, such as facial attitude, functional MRI, and electroencephalography (EEG) signal. This study used relative power spectral analysis of EEG signals to binary classify two mental states: concentration and rest. The authors also suggested a new experimental setup using the effect of Motion Induced Blindness (MIB) to stimulate and assess the concentration effectiveness. In this study, the processing block diagram can classify two mental states using basic classifier techniques such as the Support Vector Machine (SVM). As a result, this classifier model achieves an overall accuracy score of 85%. The research results show that the model can not only classify the two mental states effectively but also show some potential in predicting the concentration level.