Prediction of the Survival of Patients with Cardiac Failure by Using Soft Computing Techniques

The following paper presents a piece of work done on a relatively small dataset-with 1099 samples and 20 attributes-obtained from hospital records in Hungary. It goes to prove that by using a well tuned support vector machine model brought in better predicting results in terms of accuracy and calculation cost to a classification problem compared to an artificial neural network, random forest or the decision tree models. Next further improvements were suggested for the dataset and the preparation process as well.