Enhancing Prediction of Cardiovascular Disease using Bagging Technique

Globally, cardiovascular disease is the major cause of death; According to estimates, it takes an average of nearly 18 million lives each year according to the surveys. Recent development in the medical industry can bring a good solution to this using machine learning predictions. Individuals may suffer from elevated overweight and obesity, blood sugar, blood lipids, blood pressure, and blood glucose as a result of behavioral risk factors. Here, this research work has developed a prediction model by applying an ensemble technique on KNN, Decision Tree, Random Forest, and the Support Vector Machine (SVM) algorithms. The unique combination of these basic models with the bagging technique will create an improved model that will have an advantage over the current models in the quicker and more accurate prediction of cardiac problems, hence assisting the medical professionals. To detect the outliers and important features, Recursive Feature Elimination and Inter Quartile Range (IQR) methods are used. A web interface is used to collect user records to forecast cardiac disease.