Cardiovascular Disease (CVD) Prediction with Machine Learning (ML) K-Neighbor (KNN), Decision Tree (DT), and Random Forest (RF) Classifiers

Cardiovascular disease (CVD) has recently exceeded all other reasons of death universal in both every nation. Early detection of cardiac circumstances and ongoing therapeutic supervision by specialists can lower death rate. It’s not always possible to properly observe patients daily, and a doctor cannot deliberate with a patient for a whole day since it needs more expertise, intellect, and time. Manual methods are complex to subject when used to diagnose CVD. In this sense, Machine Learning (ML) procedures are trustworthy and effective sources to identify and classify individuals with CVDs. According to the suggested study, we used ML algorithms to recognize and forecast human CVD, and we used the CVD dataset to evaluate the act of those algorithms using various metrics, including classification accuracy, precision, F1-score and recall. In this research, we developed and researched models for CVD prediction using the patient's various heart attributes as well as CVD detection using ML techniques such as K-Neighbors Classifier (KNN), Decision Tree Classifier (DT), and Random Forest Classifier (RF) on the dataset made openly available in Kaggle website. The experiment result shows RF classifier has the highest accuracy among other classifiers. Almost it predicts 87 % of highest accuracy.