Prediction of cardiovascular disease

Nowadays, cardiovascular diseases are one of the leading causes of death. Earlier and better detection of such diseases would lead to earlier treatment and eventually to better chances of patients being able to overcome those diseases. Machine learning algorithms have been proven useful in detecting several medical conditions based on patients’ characteristics. In this paper, we are trying to predict whether a patient has a cardiovascular disease based on their characteristics. Using decision trees (C4.5), k-NN, and Naïve Bayes, in combination with cross-validation and holdout methods, we were able to achieve relatively good results. Even better results were achieved, for some specific cases such as patients having hypertension stage 2 or 3.