A COMPARATIVE ANALYSIS OF FEATURE EXTRACTION AND MACHINE LEARNING BASED CLASSIFIER FOR EEG SIGNAL CLASSIFICATION

Abstract

Electroencephalogram (EEG) is a trial did on the brain to record the electrical activity inside it. The neural structure of the brain can consist of various neurons in terms of lacs or crores. These neurons communicate by colliding among themselves and communicating data to each other. This collision leads to the generation of the very small amount of electricity. The electrical signal generated can then be recorded and carefully studied to solve many neurological disorder diseases for example epilepsy. About 1% of the total population in the world are affected by this disease. In this study, the behavior of the EEG signals was analyzed by extracting the required important features, as well as classifying the extracted features to detect epileptic seizures. This analysis was done using different machine learning techniques such as Multilayer Perceptron Neural Network (MLPNN), Radial Basis Function Neural Network (RBFNN) and Support Vector Machines (SVM).