Integrasi Gradient Boosted Trees dengan SMOTE dan Bagging untuk Deteksi Kelulusan Mahasiswa

Achmad Bisri (Universitas Pamulang)
Rinna Rachmatika (Universitas Pamulang)

Jurnal Nasional Teknik Elektro dan Teknologi Informasi
Vol 8 No 4: November 2019

Education has an important role in life. Pamulang University is a university which provides education at affordable cost. However, based on student academic performance data, there is imbalance in class between the number of students who graduate on time and students who can not graduate on time, on various study programs. In this paper, an implementation of SMOTE and bagging techniques was conducted on the Gradient Boosted Trees (GBT) classification method for handling the class imbalance problem. The proposed method is able to provide significant results with an accuracy of 80.57% and an AUC of 0.858, in the category of good classification.