Machine Learning Models to Decode Complex Patterns in Heart Disease Data: Challenges and Opportunities

An UN agency of World health organization has published a survey recently that 17.9 million individuals die annually owing to heart connected diseases and its increasing day-by-day. Nowadays, it is challenging to diagnose health issues and monitor the suitable medicine at the proper time due to the growing population and prevalence of disease. However, there is an anticipation that advancements in technological developments have accelerated up the field of general public health by spawning cutting-edge, pragmatic medical expert solutions. This work analyzes assorted supervised predictive models like Naive Bayes (NB), Decision Tree (DT), Support Vector Machine (SVM), kNN Classifier, Logistic Regression (LR), Ridge Classifier and other ensemble methods like Random Forest Classification (RF), Gradient Boosting, XGBoost Classifier and AdaBoost Classifier by employing a heart disease data that include varied predictors like gender, age, blood pressure level, blood glucose etc. The role of ensemble models on identifying and handling complex patterns in the dataset has been presented. These machine learning techniques shows varying performance and investigations are done on underlying reasons.