In this study, We evaluated comparative performance of Atrial Fibrillation outbreaking prediction model which developed three different data mining algorithm(logistic model, decision tree model, network model). First, We developed three data mining model with 1895 cardiac medical data set and evaluated each model and comparative evaluation through right and error estimating rate. We could find that the most significant factor outbreaking Atrial Fibrillation is heart rate(tachycardia state) all the three model and the best estimating model is neural network model and decision tree model. We will help medical doctor to estimating atrial fibrillation outbreak through data mining.