ABSTRACT
Psychiatry and neurology are one of the most important diagnostic approaches. Sleep specialists have taken up time consuming and challenging activities. However, pre signal processing has always been helpful in terms of accuracy. Sleep analysis is centered on a comprehensive EEG study and can only be read and interpreted by a specialist in this area. Computational intelligence approaches have shown a positive outcome in different sleep stage classifications. Sleep classification plays a valuable role as it can help identify different sleep associated disorders such as anxiety, restless leg syndrome (RLS), parasomnia, and several more. While reviewing this sleep classification concept based on EEG signals, I have categories methods in supervised learning, unsupervised learning and deep learning. Specific algorithms for all stages of sleep function differently. As every procedure has its own different classifiers. Having said that SVM, k-means and CNN are the algorithms often used and deliver decent results if not better.
