IMPLEMANTION OF SUPPORT VECTOR MECHINE AND RANDOM FOREST ALGHORITHM TO DIAGNOSE HEART DISEASE

Abstract

Heart disease is the leading cause of death in the world according to the World Health Organization (WHO). And in Indonesia, heart disease ranks as the second most deadly disease at 12.9%. Along with advances in computer technology and machine learning, software is currently evolving to assist doctors in making decisions without having to go directly to a specialist. Medical diagnosis is considered an art despite all the standardization efforts which is largely due to the fact that medical diagnosis requires expertise in dealing with uncertainties not found in today's computing machines. In this final project, a software using the Support Vector Machine algorithm with an accuracy of 93.44% and Random forest with an accuracy of 90.16% is presented which is a machine learning algorithm to assist the initial examination of patients. In diagnosing a heart attack, a detailed history or medical interview will be conducted regarding the symptoms and a physical examination. This system is intended for internal medicine, especially those with complaints of chest pain, whose function is to check early on the possibility of a heart attack. The purpose of the final project of this system is as a preventive measure so that it can detect patients who may have a heart attack early.