Selected Papers Presented at MODSIM World 2011 Conference and Expo
MODSIM World 2011 Conference and Expo, October 11-14, 2011, Virginia Beach, Virginia
In this paper, we present methods to analyze and improve an EEG-based engagement assessment approach, consisting of data preprocessing, feature extraction and engagement state classification. During data preprocessing, spikes, baseline drift and saturation caused by recording devices in EEG signals are identified and eliminated, and a wavelet based method is utilized to remove ocular and muscular artifacts in the EEG recordings. In feature extraction, power spectrum densities with 1 Hz bin are calculated as features, and these features are analyzed using the Fisher score and the one way ANOVA method. In the classification step, a committee classifier is trained based on the extracted features to assess engagement status. Finally, experiment results showed that there exist significant differences in the extracted features among different subjects, and we have implemented a feature normalization procedure to mitigate the differences and significantly improved the engagement assessment performance.
"Distribution/Availability: Unclassified - Unlimited."
Original Publication Citation
Li, F., Li, J., McKenzie, F., Zhang, G., Wang, W., Pepe, A., Xu, R., Schnell, T., Anderson, N., & Heitkamp, D. (2012) Engagement assessment using EEG signals. In Selected papers presented at MODSIM World 2011 Conference and Expo (pp. 200-207). NASA Center for Aerospace Information.
Li, Feng; Li, Jiang; McKenzie, Frederic; Zhang, Guangfan; Wang, Wei; Pepe, Aaron; Xu, Roger; Schnell, Tom; Anderson, Nick; and Heitkamp, Dean, "Engagement Assessment Using EEG Signals" (2012). Electrical & Computer Engineering Faculty Publications. 364.