A review of EEG signal classifier based on deep learning

Abstract

Electroencephalogram (EEG) signal recognition is an active research topic in the field of artificial intelligence and has been gaining extensive attention and engineering communities. This technology is an important basis of human computer interaction and many other fields. The deep learning theory has made remarkable achievements on feature extraction and gradually extended to the time sequences of EEG research. This paper reviews the traditional feature extraction of EEG recognition and discuss the traditional classification methods of EEG recognition such as linear discriminant analysis (LDA), support vector machine (SVM) and long short time memory (LSTM). Finally, this paper summarizes advantages and disadvantages of these methods. Through the feature extraction by the wavelet transform and LSTM classification, it has achieved 98% accuracy and verified that LSTM is suitable for EEG signal with time sequence feature.