Abstract
Heart disease places a significant strain on healthcare systems. Moreover, it kills millions of people each year, which is a leading cause of death worldwide. Hypertension, diabetes, unhealthy lifestyles, and genetic predispositions are all risk factors for heart disease. However, it is not easy to identify a heart disease. For this reason, helping in identifying the heart disease is important to prevent a death, e.g., caused by a heart attack. In this study, we aim to develop a heart disease prediction system. The system is developed according to the Cross-Industry Standard Process for Data Mining (CRISP-DM) framework by employing machine learning algorithms. In this work, Logistic Regression (LR) and Random Forest (RF) are utilized as our machine learning algorithms for classifying heart disease using a heart disease dataset from Kaggle. Our results show LR has an AUC value of 0.921 and F1-Score 0.89 that outperforms RF with an AUC value of 0.920 and F1-Score 0.84 in this work. Then, we select LR to be applied to the developed heart disease prediction system.