A Transfer Learning Algorithm Based on CSP Regularizations of Recorded EEG for Between-Subject Classiftcation

Feature extraction and classification are the most important parts of BCI systems. The new branch of BCI studies focuses on the design of a classifier that is trained to function properly for each individual. This problem is known as Transfer Learning. In between-subject classification, due to the differences in the neural signals’ distribution of different individuals, using the common methods of feature extraction for training the classifier, does not lead to high accuracy for the test subject. As a result, in this study, we present a method for extracting features that perform well in between subjects classifications. The data that we used in this study are EEG signals recorded during five mental tasks from nine participants with motor disabilities. The proposed method is generalized from the conventional CSP approach. To this end, we proposed a combination of two CSP regularization methods. Finally, to show the efficiency of the algorithm, we applied it to the data and compared the results with those of the previously used methods. The algorithm that we suggested led us to the best accuracy (73%) in comparison to other previous methods. Therefore, this algorithm can be a useful tool in any between subject classification problem.