Health care professionals face a challenging task of predicting and detecting heart diseases. Many hospitals have expensive mechanisms and machines in order to detect heart disease in a patient. Heart diseases have now become a common illness and needs to be diagnosed in a cost-effective way so that what health care facilities are reachable to the common man. Angiography is one of those expensive techniques that is used to detect heart disease. However, angiography is generally done at a very later stage almost just before an operation to detect the blocks in the arteries of the heart. If heart disease can be predicted at a very early stage, the patient's health can be safeguarded, and medical expenses could be reduced significantly when a patient takes appropriate steps to reduce the risk of heart disease. Some of the main and common reasons that are supposed to cause heart diseases are consuming alcohol, smoking, high calorie diets data rich in empty carbohydrates and lifestyle factors such as lack of exercise. The data generated by the healthcare industry has been studied over the years and machine learning has proved to be an effective in making decisions and predicting the results. During the study some of the algorithms such as logistic regression, decision trees, support vector machines, artificial neural networks etc have been used to create machine learning models for the heart disease data set. we have also recorded the performance of various algorithms and summarized them. We also aim to deploy a practical machine learning model using the best algorithm that suits the scenario to a web application where a user can register himself and give his inputs about his health condition so that he can get predictions a weather he's susceptible to heart disease or not.