Predicting Patterns of Student Graduation Rates Using Naive Bayes Classifier and Support Vector Machine

In Indonesia education is one of the important aspects to be implemented by anyone aiming to educate and create a reliable and resilient generation. One of the forms of education is higher education. As we know, registration data in higher education, such as student profile data, courses, KRS (Study Plan Card), alumni data, English language skills, and so on can be important information to make a policy that improves the quality of a college, and especially for a department. There is quite a large amount of this data if it has been collected for several years. This research uses data gathered, namely, student profile data, GPA, Senior High School, and residence of student to get information of our student enrollment data. By using classification methods such as Naive Bayes Classifier and Support Vector Machine, it can be used to predict whether the student graduates in a timely fashion or not. Timely graduation is defined by student graduating in four years or eight semesters, or less. Based on the research, the results obtained for this classification by using the method of Support Vector Machine are better than the Naive Bayes Classifier, with an accuracy of 69.15% for this data.