An electroencephalogram (EEG) records brain activity and gives an indication of how the brain is working throughout time. Medical personnel and researchers routinely use EEG to examine brain activity and detect neurological issues. EEG is a helpful tool to examine brain activity and is often used to identify epileptic seizures. While interpretability of methods has received comparatively less attention, most contemporary approaches to classification are primarily concerned with increasing identification accuracy. The focus of this epileptogenic classification is on legible EEG data. Comparing several categorization techniques for recognizing epileptic episodes is the goal of this research. This study aims to evaluate the efficacy of several categorization schemes in quickly identifying periods of epileptic seizures. In this work, we use several classifier methods on a dataset of seizures from epileptic patients including both black-box (SVM and KNN) and non-black-box (decision tree and random forest) classifiers. Our research shows that non-black box classifiers outperform black-box classifiers. The experimental findings show that the Random Forest classifier can help detect seizures with a high rate of accuracy. It provides excellent accuracy and low probability of misidentification compared to other classifiers. The purpose of this paper is to provide a summary of the statistical properties of machine learning and several contemporary strategies based on different classifier types.