The aim is to develop a model applying machine learning algorithms on data set obtained from ongoing cardiovascular research in Framingham, Massachusetts which has information of 4240 subjects (rows). Each of these subjects has 15 features (columns), and was classified into 0 or 1 (1 risk of CHD in 10 years and 0 no risk). All the attributes used are factors that can contribute to and pose a risk of stroke.
In 1948, more than 5,000 residents of Framingham, MA, agreed to participate in a long-term ongoing study of heart disease, to be administered by the newly established National Heart Institute (today the NHLBI) of the National Institutes of Health. [4]
The commitment of the residents of Framingham, MA, has been crucial in the study's efforts. Their cooperation has contributed to many of the major findings of heart disease made in the last half-century. Their continued participation is a crucial element for future success in the battle against the nation's number-one killer—cardiovascular disease. [4]
In this project, I have built a predictive model using Framingham Heart Study data to predict Coronary Heart Disease (CHD) in 10 years and make recommendations to prevent heart disease.
Demographic:
Behavioural
Medical (current):
Predictive variable (desired target):