Enhanced Classification Methods for the Depth of Cognitive Processing Depicted in Neural Signals
Analyzing brain states is a difficult problem due to high variability between subjects and trials, therefore improved techniques are requested to be developed for a better discrimination between the neural components. This paper investigates multiple enhanced classification methods for neurological feature selection and discrimination of the depth of cognitive processing. The aim is to detect the strengths and weaknesses of different classification methods and benefit from their highest performances, so that the neural information could optimally be detected. As a result, we obtained a classification rate improved by at least 5% by integrating complementary information that better describe the neural activity.
Published in: Scientific Bulletin: Series C, Electrical Engineering and Computer Science, University Politehnica of Bucharest