Minimizing Attributes for Prediction of Cardiovascular Diseases

This study is aimed at the early detection of cardiovascular diseases using predictions learning with a high percentage of successes using the lowest possible number of attributes. Results are comparable to other techniques. Applying the learning system through the reduction of attributes, a tree was obtained with the same classification result of 85.8% that appears in the literature, but it was obtained with 5 variables using the decision tables technique (Decisions Table) and Bayesian Networks.