| dc.contributor.advisor | Syafitri, Utami Dyah | |
| dc.contributor.advisor | Erfiani | |
| dc.contributor.author | Zidan, Muhamad Fawaz | |
| dc.date.accessioned | 2025-07-10T04:42:39Z | |
| dc.date.available | 2025-07-10T04:42:39Z | |
| dc.date.issued | 2025 | |
| dc.identifier.uri | http://repository.ipb.ac.id/handle/123456789/164484 | |
| dc.description.abstract | Penelitian ini bertujuan untuk mengidentifikasi faktor-faktor yang memengaruhi Indeks Prestasi (IP) mahasiswa tahun pertama program D4 IPB University angkatan 2024. Mahasiswa ini masuk melalui berbagai jalur penerimaan yang diatur dalam Peraturan Menteri Pendidikan Nomor 48 Tahun 2022. Setelah menyelesaikan semester pertama, mahasiswa memperoleh IP yang menjadi dasar penentuan beban studi pada semester-semester berikutnya. Penelitian sebelumnya menunjukkan bahwa latar belakang sekolah dan jalur penerimaan berpengaruh signifikan terhadap performa akademik. Penelitian ini menerapkan metode ensemble learning, yaitu Random Forest dan Extreme Gradient Boosting (XGBoost), yang dapat meningkatkan akurasi prediksi dan meminimalkan overfitting. Data diperoleh dari Direktorat Administrasi Pendidikan dan Penerimaan Mahasiswa Baru (DAPPMB) IPB tahun 2024. Data diproses menggunakan teknik klasifikasi dengan cross-validation, serta dievaluasi menggunakan metrik evaluasi, yaitu accuracy, precision, recall, dan F1-Score. Tiap metode dimodelkan dengan dan tanpa menggunakan class weighting serta hyperparameter untuk hasil yang lebih optimal. F1-Score model terbaik dari metode Random Forest adalah 0.644 yang diperoleh tanpa class weighting, sedangkan XGBoost adalah 0.431 yang diperoleh tanpa class weighting. Hasil penelitian menunjukkan bahwa model Random Forest tanpa class weighting memberikan performa terbaik dibandingkan model lainnya, dengan F1-Score tertinggi sebesar 0.644. Peubah berpengaruh dipilih dari model terbaik, yaitu Random Forest tanpa class weighting. Peubah yang berpengaruh signifikan terhadap indeks prestasi mahasiswa D4 IPB University tahun 2024, antara lain rataan total rapor, rataan syarat rapor, dan program studi diterima. | |
| dc.description.abstract | This study aims to identify the factors influencing the Grade Point Average (GPA) of first-year D4 students at IPB University in the 2024 cohort. These students admitted through various admission pathways regulated under the Ministry of Education Regulation No. 48 of 2022. After completing their first semester, students receive a GPA that is the basis for determining their academic load in subsequent semesters. Previous studies have shown that school background and admission route significantly affect academic performance. This research applied ensemble learning methods, namely Random Forest and Extreme Gradient Boosting (XGBoost), known for enhancing prediction accuracy and minimizing overfitting. The data were obtained from the Directorate of Academic Administration and New Student Admission (DAPPMB) of IPB in 2024. The data were processed using classification techniques with cross-validation and evaluated using performance metrics such as accuracy, precision, recall, and F1-Score. Each method was modeled both with and without class weighting and hyperparameter tuning to achieve more optimal results. The best F1-Score from the Random Forest method was 0.644, obtained without class weighting, while XGBoost achieved its best F1-Score of 0.431, also without class weighting. The results indicate that the Random Forest model without class weighting provided the best performance among all models, with the highest F1-Score of 0.644. Influential variables were selected based on this best-performing model, that is Random Forest without class weighting. The variables that significantly influenced the GPA of D4 students at IPB University in 2024 include the average total report card score, average requirement report card score, and accepted study program. | |
| dc.description.sponsorship | | |
| dc.language.iso | id | |
| dc.publisher | IPB University | id |
| dc.title | Identifikasi Peubah Berpengaruh terhadap Indeks Prestasi Mahasiswa SNBP Diploma IV IPB University Tahun 2024 dengan Metode Random Forest dan XGBoost | id |
| dc.title.alternative | Identification of Influential Variables on the Grade Point Average of SNBP Diploma IV Students at IPB University in 2024 Using Random Forest and XGBoost Methods | |
| dc.type | Skripsi | |
| dc.subject.keyword | ensemble learning | id |
| dc.subject.keyword | indeks prestasi mahasiswa | id |
| dc.subject.keyword | feature importance | id |
| dc.subject.keyword | random forest | id |
| dc.subject.keyword | XGBoost | id |
| dc.subject.keyword | grade point | id |
| dc.subject.keyword | peubah berpengaruh | id |