Development of automatic detection methods for identifying different heart irregularities or arrhythmias was required due to the rising number of heart patients in order to ease the burden on doctors. An electrocardiogram (ECG) records the heart's electrical activity and can be used to detect heart problems. ECG signals are widely utilized to categories the signals into several classes that aid medical professionals in identifying heart disorders. It is very challenging to classify an ECG accurately. In recent years, various techniques have been employed and investigated for the classification of cardiac signals and heart rhythm problems. The classification of ECG signals often uses temporal, morphological, fast Fourier transform, wavelet features, statistical features, correlation, and regression techniques. In this study, strategies for extracting and analyzing ECG signals are reviewed. It has been found that a hybrid feature extraction improves detection effectiveness. The majority of authors have worked on to classify ECG into 5 categories. Therefore, there is room to find the elements that work best together to deliver optimal effectiveness for a larger number of ECG classes. Future work is possible with a classifier that can deliver the highest accuracy with a larger number of classifications.