Prediksi Tingkat Kelulusan Mahasiswa menggunakan Algoritma Naïve Bayes, Decision Tree, ANN, KNN, dan SVM

Abstract

The student graduation rate in all universities can be measured by looking at their study duration, both on time and delayed. Thus, by observing the study duration, it can affect the quality of study programs in universities. The purpose of this research is to apply and compare the Naïve Bayes, Decision Tree, Artificial Neural Network, K-Nearest Neighbor (K-NN), and Support Vector Machine (SVM) algorithms in predicting the graduation rate of students. The dataset in this research consisted of 807 student data from the Faculty of Engineering, Universitas Hamzanwadi. The data analysis technique used was descriptive statistics by applying the knowledge discovery in a database (KDD) method. The testing of the five algorithms was done by optimizing the data using the SMOTEENN technique, with a data split of 80% for training and 20% for testing, using a random state of 42. Our findings show that the Naïve Bayes algorithm had an accuracy of 92.37%, Decision Tree 91.60%, K-NN 96.95%, SVM 93.13%, and ANN 90.84%. Among the five algorithms tested, the K-NN algorithm had the highest accuracy rate of 96.95%. The predicted results tended to indicate delayed graduation influenced by GPA. Therefore, the institution needs to pay more attention to students predicted to be delayed to improve their GPA each semester, thus promoting timely graduation within the expected time frame.