IoT based heart disease prediction using smote and machine learning techniques Available to Purchase

Cardiovascular Disease (CVD) or cardiac arrest refers to several diseases affecting the heart and has become a leading cause of death in the last few decades. In the health industry, machine learning (ML) improved the efficiency of predictive models. Using the Internet of Things (IoT) makes it possible to predict appropriate medical treatments and lifestyle changes early on. This paper describes supervised machine learning classifiers for detecting significant features which predict cardiovascular disease with great accuracy. Over-sampling is a method for learning from unbalanced data that we propose in this work. Different classification techniques are incorporated into our prediction model. An analysis of K-Nearest Neighbor, Naive Bayes, Logistic Regression, and Decision Trees techniques are performed to determine which ML approach performs the best on heart disease datasets. The accuracy percentage for the Decision Tree using SMOTE oversampling technique is 91.26, this predicts heart disease with the maximum probability.