Ordinary Learning Method for Heart Disease Detection using Clinical Data

Heart diseases are one of the major causes of human deaths today. About 610000 human beings expire annually in the United States due to this fatal disease and the condition is more severe in the underdeveloped countries lacking medical experts. Accurate detection of heart disease in a human being can be helpful in proper medication against this lethal disease and considerably reduce this alarming death rate. Data mining and machine learning techniques are being widely used for medical diagnosis these days. This research paper employs Ordinary Learning Method for the accurate detection of heart disease using clinical data. The proposed method is tested on the Standard UCI(University of California, Irvine) Cleveland Heart Disease dataset using 14 attributes. The achieved accuracy of the proposed method is 98.4615% which is compared with other states of the art techniques such as C5.0 decision trees, Support vector machine, KNN and Neural Network. Comparison results show that the proposed OLM technique outperforms the previous data mining techniques proposed in literature for the detection of heart disease.