부산대학교 도서관

Heart failure (HF) is one of the most frequent illnesses in modern times, and it can lead to grave scenarios. About 26 million people suffer from this disease every year. From the point of view of cardiology consultants and surgeons, predicting heart failure with improved accuracy is challenging. There are various classification and predictive models which can assist the healthcare sector and explain how to efficiently use data for effective predictions. To that goal, this research study has applied a range of machine learning (ML) algorithms to comprehend the data and predict the occurrence of heart failure in medical databases. Additionally, the findings and comparison analysis determined that this study improved on the prior assessment of the accuracy of predicting heart disease based on the prominent attributes. This study’s combination of machine learning models with medical information systems can assist in predicting the heart failure and other disorders using data obtained from the patients. The implication of k-fold cross validation for predictive model building and assessment insisted that the performance of the model could be enhanced for more reliable predictions. The classifiers used in this study are-Decision Tree, Random Forest, and Logistic Regression, in which Decision Tree provided more accurate results than the other classifiers.