The graduation rate on time in college cannot be taken lightly. In many cases, it is found that there is an imbalance in the number of students who enter and who have completed their studies so that there is a high accumulation of the number of students in each batch. It is necessary to know the factors that cause students not to graduate on time. Data classification techniques can be used to predict the timeliness of student graduation. The algorithm used is the Naïve Bayes algorithm with data from 122 students from the Informatics Engineering and Information Systems study program at the Faculty of Computer Science, AMIKOM University, Yogyakarta. Based on the results that have been carried out, it can be concluded that the Naïve Bayes algorithm is applied accurately to predict the timeliness of students' graduation. The results of testing the level of prediction accuracy resulting from the trial reached 90.16%.