Utilizing Fuzzy-SVM and a Subject Database to Reduce the Calibration Time of P300-Based BCI

Current Brain-Computer Interfaces (BCI) suffer the requirement of a subject-specific calibration process due to variations in EEG responses across different subjects. Additionally, the duration of the calibration process should be long enough to sufficiently sample high dimensional feature spaces. In this study, we proposed a method based on Fuzzy Support Vector Machines (Fuzzy-SVM) and a database of training samples from several subjects to address both issues for P300-based BCI. To validate the proposed approach, we conducted P300 speller experiments on 18 subjects and formed a subject-database using the leave-one-out approach. Fuzzy-SVM is an extension to the traditional SVM in which a different weight is assigned to every slack variable. We assigned the same weight to all the slack variables coming from a specific subject in the database. The weight of a subject in the database set to be proportional to the accuracy obtained by a standard SVM which is trained using only samples from the corresponding subject and tested with samples of the test-subject. With the proposed approach, we achieved to obtain an average accuracy of 80% with only 4 training letters. Conventional subject-specific calibration approach, on the other hand, needed 12 training letters to provide the same performance.