Publication

Abstract : In recent years, heart disease has adversely affected people’s quality of life. Since the mortality rate from heart disease is still relatively high, there is a need to increase efforts in prevention to improve the prediction model for heart disease. Machine learning (ML) has accomplished excellent results in decision-making and future prediction using the enormous amount of data generated by the healthcare sector. In this paper, with feature analysis new features like body mass index (BMI) and mean arterial pressure (MAP) are created in the prediction process. Our experiments in this paper are carried out using a variety of ML classification methods, namely Naive Bayes, Logistic Regression (LR), XG Boost, LG Boost, Ada Boost, and Stochastic gradient descent. Enhancement of the system’s performance is done by using ML classification techniques with HyperOpt parameter tuning technique. The model is evaluated using a Kaggle heart disease data set, which contains over 70,000 records. According to the experimental results, XG Boost and LG Boost with the HyperOpt achieved the highest level of accuracy.