<![CDATA[Student admissions in universities every year become a routine thing to do, some even do student admissions every semester. That way, the number of students will continue to grow. Especially if there are students who graduate late, it will increase the number of students in the university. There are many things that can affect graduation, namely personal data (gender, age, marital status, job status) and academic data (grade). Before making a decision, universities must analyze the number of students and the factors that most influence student graduation. Analysis by classifying graduation using C4.5 algorithms. The research method used consists of selection to ensure the data used in the KDD process is appropriate and quality data. Then preprocessing by means of data cleaning, data reduction, and data normalization. The next method is transformation for age attributes to young and old, grade attributes to large and small. The last method is C4.5 algorithm modeling with rapid miner and evaluation. Through the calculation process using the classification method and C4.5 algorithm with the attributes described earlier, the results were obtained that the accuracy of the graduation classification reached 97.00%, the precision value was 91.79%, and the recall value was 100.00%, and the AUC value was 0.978. This means that the model has a very high level of accuracy and has an excellent ability to separate samples from both target classes.]]>
Copyrights © 2024
