EEG Feature Fusion for Person Identification Using Efficient Machine Learning Approach

Abstract

Recently, the electroencephalogram (EEG) signal presents an excellent potential for a new person identification technique. This paper proposed a new method for EEG feature extraction based on fusing different EEG features. In general, EEG feature extraction can be categorized into three types which are time domain, frequency domain, and time-frequency domain features. This paper also applied several supervised learning approaches to select the efficient classifier for EEG-based person identification. The performance of the proposed method is tested using standard EEG datasets, namely, EEG Motor Movement/Imagery Dataset. The results are evaluated using four common criteria which are: accuracy rate (ACCEEC), sensitivity (SenEEC), specificity (SpeEEC) and F-score (FSEEC). The experiment results show that the fusion approach achieves better results compared with a traditional EEG feature extraction approach. The proposed fusion feature method is recommended to apply in more challenging signal problem instances, such as user authentication or early detection of epilepsy based on EEG signals.