Distributed classifier migration in xcs for classification of electroencephalographic signals

TL;DR: Results indicate that classifier migration is an effective method for improving classification accuracy, improving learning speed and reducing final classifier population size, in the single-step classification of noisy, artefact- inclusive human electroencephalographic signals.

Abstract: This paper presents an investigation into combining migration strategies inspired by multi-deme parallel genetic algorithms with the XCS learning classifier system to provide parallel and distributed classifier migration. Migrations occur between distributed XCS classifier sub-populations using classifiers ranked according to numerosity, fitness or randomly selected. The influence of the degree-of-connectivity introduced by fully-connected, bi-directional ring and uni-directional ring topologies is examined. Results indicate that classifier migration is an effective method for improving classification accuracy, improving learning speed and reducing final classifier population size, in the single-step classification of noisy, artefact- inclusive human electroencephalographic signals. The experimental results will be used as part of our larger research effort investigating the feasibility of using EEG signals as an interface to allow paralysed persons to control a powered wheelchair or other devices.