A Review on Classification Methods Used in EEG-Based Home Control Systems

Abstract

Brain-Computer Interface (BCI)-based system enables a subject to communicate to a computer machine without performing any muscular activity. In the BCI-based system, the user can control home appliances like television, fan, door, light system etc. without any physical movement. This is possible by converting the classified signal into command. The various studies of transition control command using BCI (e.g. robot and gaming control etc.) have already been achieved. BCI gives a platform for the physically disabled people to interact with the environment and makes their life better. This paper reviews the classifier used for the EEG-based home automation system. The reviewed classification method involves linear discriminant analysis (LDA), random forest, adaptive neuro-fuzzy inference system (ANFIS), artificial neural network (ANN) etc. The classifiers are compared on the basis of accuracy. Based upon accuracy, LDA as a classifier gives good accuracy for the BCI-based home control system.