Cardiovascular diseases (CVDs) are the world's leading cause of mortality. The current method of diagnosing the disease is the analysis of an electrocardiogram (ECG). The physicians find it difficult to accurately diagnose abnormal heart behavior. However, early and precise detection of cardiac abnormalities helps in providing appropriate treatment to patients. The development of automated ECG classification is an emerging tool in medical diagnosis for effective treatment. In this paper, an effective technique based on Artificial Neural Networks (ANN) is described to classify ECG data into two classes: normal and abnormal. In this context, ECG data are obtained from UCI Arrhythmia databases where the classification is conducted using MATLAB platform. The experimental findings demonstrate that the proposed technique achieves a high classification accuracy of 92.477%, allowing it to effectively detect ECG signal abnormalities and implement it to diagnose heart disease.