Hepatitis is an inflammation disease of the liver because infection that attacks and causes damage to cells and liver function. Hepatitis is a disease precursor of liver cancer. Hepatitis can damage liver function as neutralizing poisons and digestive system in the body that break down nutrients and then spread to all organs of the body that very important for humans. Research of predicting disease hepatitis have been carried out by previous researchers. This research using the method of classification data mining algorithm C4.5 and Naïve Bayes is then performed comparative to both methods., The measurement of two method using cross validation, confusion matrix and ROC curve. The result of this research is the best algorithm that can be used to predict disease hepatitis.