Heart arrhythmia is a state of the heart in which the heartbeat is unbalanced which can be too fast, too slow, or unstable. Electrocardiography (ECG) is used for the recognition of Heart arrhythmia. It registers the electrical activities of the heart of a patient for a period with electrodes attached to the skin. Due to the ECG signals that reflect the physiological conditions of the heart, medical specialists tend to utilize ECG signals to detect and analyze heart arrhythmia. The most important skill of medical doctors is being able to identify the dangerous types of heart arrhythmia from ECG signals. In spite of this, interpretation of the ECG waveforms performed by professional medical doctors manually is proven to be monotonous and time-consuming. As an end result, it was found that the development of automatic systems for identifying abnormal conditions from diurnal recorded ECG data is of primary importance. In addition to that, suitable timely medical treatment measures can be effectively applied when such irregular heart conditions can be identified instantly using health monitoring equipment and tools which internally use machine learning algorithms. Therefore, an important role in this regard would be machine learning.