Epilepsy identification is done by visual observation of electroencephalography (EEG) signals, which is more sensitive to bias and time consuming. In most of the previous research of epileptic seizure detection suffers from unsuitability and low power for processing large datasets. To eliminate aforementioned problems a computerized detection method is required to aid medical professionals. In this paper, a new technique is proposed to identify the epilepsy based on VMD, RELIEFF algorithm and machine learning approach. To investigate the proposed method performance a public EEG dataset is adopted from university hospital bonn, Germany. The technique starts with the VMD, which is used to extract the features from each EEG signal. And then RELIEFF algorithm is adopted to identify the best features. Finally to categorize the normal and epilepsy EEG signals a machine learning classification (ANN, KNN, and SVM) approach is used. The results demonstrate that the adopted method (VMD+RELIEFF+SVM) can achieve a better accuracy, shows that a commanding method to identification and classification of epileptic seizures
Classification of EEG Signals for Epileptic Seizures Using Linear and Non-linear Classifiers Based Wavelet Transforms and Information Criteria
