Electrocardiogram (ECG) is a painless and noninvasive way to help diagnose many common heart problems in people of all ages. It is a very useful tool for diagnosing any cardiovascular disease such as arrhythmia. Arrhythmia refers to any change in the pattern of ECG signals and also the change of beat rate. That means arrhythmia sums up to most of heart diseases. The beat rate may get faster or slower than the regular. According to the statistics, 80% of sudden cardiac arrest may lead to death [8]. In modern computer science, ECG classification may produce some problems [9]. Feature extraction from the dataset can have lack of good distribution and it may collect temporary feature selection which may causes a bad classification. ECG beats can vary with timing, amplitude of the signals coming from different patients. Each ECG pattern has similar beat shape. Without proper data processing, the classification may give a poor performance. Human emotions can change the heart beat shape such as feeling excited might race the heart beat faster than usual. So, every classification might get some issues regarding the real time data analysis and feature extraction.