In order to facilitate the learning process, Universitas Dinamika Bangsa (UNAMA) has a database. Every year, the alumni data gets larger, and the database can be used. Utilising alumni data involves classifying and analysing long-term study periods of Universitas Dinamika Bangsa (UNAMA) students using the naïve bayes method.The results of the naïve bayes classification in the student of information system with the highest accuracy are obtained by using the Use Training Set, which consists of 161 correctly classified instances and 39 incorrectly classified instances, with an accuracy percentage of 85% for correctly classified instances and 19.5% for incorrectly classified instances. The results of attribute selection using the Classifier Attribute Evaluation algorithm (ClassifierAttributeEval) indicate that IPK is the attribute that has the greatest influence on kelulusan speed. Akurasi in the model is calculated using a confusion matrix, and at the beginning of the data-data mahasiswa, there is a lot of noise, which is revealed through the data cleaning process. It is the process of reducing noise in data using Microsoft Excel, which the author typically uses to analyse data From of information System students. Overall accuracy is 77.5%, which is a very good accuracy when analysing training data.