Evolutionary feature selection and electrode reduction for EEG classification

摘要: EEG signals usually have a high dimensionality which makes it difficult for classifiers to learn the difference of various classes in underlying pattern signal. This paper investigates several evolutionary algorithms used reduce data. The study presents electrode and feature reduction methods based on Genetic Algorithms (GA) Particle Swarm Optimization (PSO). Evolution-based are generate set indexes presenting either seats or points that maximizes output weak classifier. results interpreted achieved, significance lost accuracy, possibility improving accuracy by passing chosen electrode/feature sets alternative classifiers.