Unusual spikes in electrical activity in the brain are a hallmark of epilepsy. Involuntary muscle contractions brought on by these seizures may cause social challenges for the person experiencing them as well as their family. Seizures may cause fainting or even death in severe circumstances. The purpose of this study is to suggest a method for accurately classifying and identifying epileptic episodes early on. The goal is to complete this task quickly and with the least amount of computing cost. We used the University of Bonn database for this study. After dividing the brain signal into four halves, we were able to extract a number of statistical features. Additionally, we extracted another set of characteristics using the Gray Level Co-occurrence Matrix (GLCM). Classifiers were then used to aggregate and classify all of the retrieved characteristics. The SVM classifier reached 100% in just 0.07 s, achieving ideal levels of sensitivity, specificity, and accuracy. Additionally, we evaluated a convolutional neural network's (CNN) performance and compared it to our suggested method. The CNN took more than three minutes to process, but it achieved 99.99% accuracy. By contrast, our suggested approach maintained the lowest possible computing cost while demonstrating higher performance comparable to deep learning networks.