This paper compares the performance of machine learning algorithms trained and tested on single-subject EEG data compared to nine-person cross-subject EEG data from the BCI IV 2a dataset. To compare the performance of single-subject and cross-subject EEG models, we implement eight machine learning algorithms and test them on EEG motor imagery data. Single-subject models had higher average accuracies compared to cross-subject trained models for 7 out of 8 machine learning models.
CITATION STYLE
Geraghty, J., & Schoettle, G. (2022). Single-Subject vs. Cross-Subject Motor Imagery Models. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 13519 LNCS, pp. 442–452). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-17618-0_31
Mendeley helps you to discover research relevant for your work.