World Health Organization estimates that 50 million people suffer from epilepsy, a neurological disorder. While electroencephalography (EEG) is useful for monitoring and diagnosing epilepsy patients' brain activity, it requires the analysis of a trained professional in order to identify any signs of epileptic activity in the recordings. Obviously, this is a slow and laborious approach. The primary goal of this investigate is to progress a perfect that can foretell the nature of an oncoming seizure.The main focus of the projected approaches is to enhance classification accuracy of epileptic seizure signals and the prediction rate. And also in this study, we projected scheme is based on Convolutional Neural Network-(CNN) with deep learning for classification. This procedure inspired by biological neural networks and are used in statistics and cognitive science. In this project we find the five kind of Seizures as Absence Seizures, Myoclonic seizures, Tonic seizure, Colonic seizure and Atonic Seizures.