One of the most commonly used instruments for the diagnosis and evaluation of epilepsy is the electroencephalogram (EEG). Currently, epilepsy be diagnosed mostly by a neurological specialist via visual or manual EEG examination readings. This study proposes an epilepsy computer-aided diagnostics (CAD) based on the Feed-Forward Neural Network (FFNN), Discrete Wavelet Transform (DWT), and Shannon entropy. DWT divides EEG impulses into numerous sub-bands of frequency that consist of gamma, beta, alpha, theta, and delta. Shannon entropy extracts ECG information from every frequency sub-band. Lastly, FFNN uses the collected features to classify the related EEG signals as normal or epileptic. The outcomes of experiment with the accessible to the public Bonn University EEG dataset indicate the total precision.