C4.5 classification is the widely used Machine Learning algorithm in variety of applications. It is a type of Decision Tree classification algorithm used by researchers to predict the future by analyzing past historical data. In this paper, C4.5 classification algorithm is implemented to calculate the survival rate of life time of a Patient affected with heart attack. The echocardiogram dataset from UCI source with 132 instances and 12 attributes were used for the experimental analysis. This dataset contains missing values which may affect the prediction accuracy and they are replaced using binning method to overcome the drawback. The implementation of the echocardiogram dataset with C4.5 classification was done with four different modes and the accuracies were measured. Mode-1: Applying Echocardiogram Dataset with Missing Values to C4.5 Classification Algorithm. Mode-2: Applying Echocardiogram Dataset after removing Missing Values to C4.5 classification Algorithms. Mode-3: Applying Echocardiogram Dataset with removed missing values to C4.5 algorithm by considering Node splitting criterion only and Mode-4: Echocardiogram Dataset with removed missing values to the C4.5 Algorithm by considering Node splitting and tree Pruning. The accuracies and other related metrics were measured to predict the possible survival rate of heart attack patients at different modes so that the patients can pre-determine whether or not to undergo expensive treatments. The performance of Mode-4 is better than the earlier method which proves that attribute split criterion and tree pruning improves the accuracy. © RJPT All right reserved.