Epilepsy is a common chronic nervous disorder affecting fifty million individuals worldwide. Epileptic Seizures are the results of the fleeting and surprising electric phenomenon of the brain. Electroencephalogram (EEG) signal is a vital information supply in diagnosing epilepsy because it records electrical and neural activities from the brain. Traditionally, these graphical record signals (EEG) are manually observed by medical practitioners which is time taking and gives improper result. In this paper, we have a tendency to propose a brand-new computer based mostly automatic epilepsy seizure detection concept. The EEG signal from epileptic and healthy patients were divided and broken into frequency sub-band with the help of Discrete Wavelet Transform (DWT). We applied feature extraction in these frequency sub-bands and extracted 10 features out of them. The features extracted were then fed into four different classifiers particularly, Support Vector Machine (SVM) k-nearest neighbor (kNN), Naïve-Bayes and Decision Tree classifiers. Six parameters are further used to compare the performance of these classifiers. An accuracy of 99.33% has been achieved in our work. © 2021, Springer Nature Singapore Pte Ltd.