Cardiovascular diseases (CVDs) are the leading cause of death worldwide and a major public health concern, with heart diseases being the most prevalent ones, thus the early prediction is being considered as one of the most effective measures for CVDs control. The risk evaluation for CVD occurrence on participants (men and women) especially aged older than 50 years with the aid of Machine Learning (ML) models is the main purpose of this research paper. The performance of supervised ML models is compared in terms of accuracy, sensitivity (or recall) in identifying those participants that actually suffer from a CVD and Area Under Curve (AUC) score. The experimental analysis demonstrated that the Logistic Regression classifier is the most appropriate against Naive Bayes, Support Vector Machine (SVM) and Random Forest with 72.1% accuracy, recall and 78.4% AUC.