This report represents the mini-project given to the students of the seventh semester for the partial realization of COMP 484, Machine Learning, provided by the department of computer science and engineering, KU. Cardiovascular diseases have been the most common cause of death worldwide in recentdecades in developed, underdeveloped and developing countries. Early diagnosis of heart diseaseand ongoing medical monitoring can reduce the deathrate. However, it is not possible to track patients every day in all cases accurately and consulting a patient for 24 hours with a doctor is not available as itrequires more wisdom, time and expertise. In thisproject, we developed and researched models to predict heart disease through different patient heart attributes and detect impending heart disease using machine learning techniques such as backward elimination algorithm, logistic regression and REFCV on the dataset publicly available on Kaggle Webs ite, further evaluate the results using confounding matrix and cross-validation. An early prognosis of cardiovascular diseases can help in making decisions about lifestyle changes in high- riskpatients and therefore reduce complications, which can be a milestone in the field of medicine.

Machine Learning, Logistic regression, Cross-validation, Back-elimination.