Max Depth Impact on Heart Disease Classification: Decision Tree and Random Forest

Abstract

Heart disease classification is an important aspect of prevention and early treatment. Heart disease classification results that are inaccurate and have low accuracy can endanger the patient's life. Several classification techniques using machine learning for heart disease have been carried out. However, there are still few studies that analyze the parameters in the algorithm model. Using inappropriate parameters can result in low accuracy. This study compares Decision Tree and Random Forest algorithms for heart disease. The max depth parameter is the parameter analyzed in this study. If the max depth is not set properly, the classification results can be inaccurate and lead to incorrect diagnoses. This study uses a holdout validation scheme for data sharing and tests different max depth parameters, namely max depth = 3, 4, 5, 6, and 7. The analysis results show that the max depth parameter that produces the best accuracy is max depth = 7 with the best accuracy result by Random Forest which is 99.29% while the Decision Tree accuracy is 98.05%. In future research, research can be conducted on the effect of other parameters by testing using several data sets.