Ensemble SWLDA Classifiers for the P300 Speller

The P300 Speller has proven to be an effective paradigm for braincomputer interface (BCI) communication. Using this paradigm, studies have shown that a simple linear classifier can perform as well as more complex nonlinear classifiers. Several studies have examined methods such as Fisher’s Linear Discriminant (FLD), Stepwise Linear Discriminant Analysis (SWLDA), and Support Vector Machines (SVM) for training a linear classifier in this context. Overall, the results indicate marginal performance differences between classifiers trained using these methods. It has been shown that, by using an ensemble of linear classifiers trained on independent data, performance can be further improved because this scheme can better compensate for response variability. The present study evaluates several offline implementations of ensemble SWLDA classifiers for the P300 speller and compares the results to a single SWLDA classifier for seven able-bodied subjects.