EEG based Mental Workload Assessment via a Hybrid Classifier of Extreme Learning Machine and Support Vector Machine

Abstract

Machine learning and neural networks are increasingly used for mental workload classification. The EEG signals can be used as an important indicator to assess the variation of the mental workload. By extracting the features of the EEG signals, the salient information can be extracted in the temporal and frequency domains. In this work, we applied a hybrid mental workload classification framework that combines extreme learning machine (ELM) and the support vector machine (SVM). The former is used as the member classifier to find hidden information in high-dimensional EEG features. The latter is used to fuse the outputs of the member classifier. Finally, we compare the proposed ELM-SVM model to classical mental workload classifiers and demonstrate its effectiveness.