Classifying student academic performance: C4.5 and SVM Methods in ITTP's Computer Engineering Program

Edelin Gultom (Institut Teknologi Telkom Purwokerto, Indonesia)
Yohani Setiya Rafika Nur (Institut Teknologi Telkom Purwokerto, Indonesia)
Rima Dias Ramadhani (Institut Teknologi Telkom Purwokerto, Indonesia)

Jurnal Mantik
Vol. 7 No. 4 (2024): February: Manajemen, Teknologi Informatika dan Komunikasi (Mantik)

The decline in the number of graduates from the Computer Engineering Program, based on the graduation percentages of the 2017 and 2018 cohorts, coupled with the imbalance between on-time and delayed graduates, poses various challenges. These challenges include suboptimal program accreditation and an excessive number of active students. This research aims to develop a classification model for student performance categorization into On-Time Graduation (LTW) and Delayed Graduation (LTTW) classes using SVM and C4.5 algorithms. The C4.5 algorithm will handle attribute selection, while SVM will be responsible for building the prediction model. The classification results will be visualized on a website using the Flask framework, allowing users to input relevant data. The classification accuracy on the test set, reaching 77.74%, indicates the model's precision in predicting student performance categories