Prediction-of-Mortality-Due-to-Cardio-Vascular-Diseases-Using-Machine-Learning-

Cardiovascular diseases are one of the leading causes of death with a global death rate of 17 million people every year with the common them being heart diseases. This typically occurs as a result of blockage or reduction in the flow of oxygenated blood to the heart. Since it is caused by several risk factors including high blood pressure and smoking, early diagnosis and disease detection is very important in reducing the effect of heart disease and inadvertently saving lives. However, modern technology has been able to significantly slow this trend in recent times through advanced research especially in the area of disease prediction and diagnosis, prognosis assessment or outcome of ailment and treatment recommendation for these specific diseases. The purpose of this project is to accurately predict mortality by heart failure through the application of a set of Gaussian Process models with different Kernel functions on a dataset containing medical information on 299 patients. The results will then be compared to a Random Forest and an XGBoost ensemble classifiers. Dataset is publicly available on Kaggle.