Performance Analysis of Heart Disorder Prediction Using Machine Learning Approaches

Abstract

Machine learning, Data mining are fundamental in health care also in health care information and identification are essential. Machine learning approaches has recently been utilized to detect and forecast a variety of major health hazards, including diabetes prediction, brain tumor detection, renal problem prediction, and Covid-19 identification, among others. The part of heart is precious organ of our body and if it has any problem then the impact is more dangerous to our body. According to the Centers for Disorder Control and Prevention (CDC) Trusted Source, heart disorder is the leading cause of death worldwide. We use a few attributes to check our heart disorder analysis, and this attribute is one of the most common causes of heart disorder. As a consequence, 6 machine learning classifiers are employed to evaluate the data using Google Collaboratory: Naive Bayes (NB), Logistic Regression (LG), K Nearest Neighbor, Bagging, Decision Tree (DT), and Random Forest (RF). Using the Seaborn distplot, we extract all attributes' features. Here, Applying Random Forest Algorithm (RF) we get the best accuracy, which is 99.18 %. We have the biggest value of the ROC (receiver operating characteristic) curve of any other algorithm.