Rahma, Nabilah Atika (2026) Pengembangan Platform Berbasis Web Untuk Analisis Dan Interpretasi Faktor Risiko Burnout Akademik Menggunakan Machine Learning Dengan Pendekatan Explainable AI. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 5025221005-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
5025221005-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (10MB) | Request a copy |
Abstract
Burnout akademik akibat tekanan belajar berkepanjangan berdampak pada performa dan kesejahteraan mahasiswa, namun identifikasinya masih banyak mengandalkan cara manual yang kurang mampu menangkap pola antarfaktor risiko secara komprehensif. Penelitian ini mengembangkan platform berbasis web untuk analisis dan interpretasi faktor risiko burnout akademik menggunakan machine learning dengan pendekatan Explainable Artificial Intelligence (XAI). Menggunakan dataset Student Stress Factors (1.100 data, 20 fitur, tiga kelas tingkat stres), penelitian menerapkan feature engineering berbasis MBI-SS, seleksi fitur, dan reduksi dimensi melalui delapan konfigurasi preprocessing, lalu mengevaluasi sebelas algoritma beserta model ensemble dengan F1-macro sebagai metrik utama. Hasil pengujian menunjukkan bahwa tahapan preprocessing, khususnya rekayasa fitur dan normalisasi, berpengaruh nyata terhadap performa model, dengan kinerja terbaik diperoleh Random Forest pada konfigurasi C4_FE_Norm (Overall Score 92,58%; Accuracy 89,55%; Brier Score 0,0396). Analisis SHAP berperan sebagai mekanisme penelusuran balik (trackback) yang menjelaskan hasil prediksi secara global maupun lokal sekaligus menjadi dasar evaluasi kualitas variabel, sehingga mengatasi black-box problem tanpa mengorbankan akurasi model secara signifikan. Model final diimplementasikan ke dalam platform berbasis Streamlit sebagai alat skrining dini yang keluarannya diverifikasi psikolog. Evaluasi usability menghasilkan rata-rata SUS 82,24 (kategori excellent) dan SEQ 6,15 dari 7, menunjukkan platform yang akurat, transparan, dan mudah digunakan.
======================================================================================================================================
Academic burnout caused by prolonged academic pressure negatively affects students' academic performance and psychological well-being. However, existing identification methods still rely heavily on manual assessments that cannot comprehensively capture the complex relationships among multiple risk factors. This study develops a web-based platform for early academic burnout risk screening using machine learning and Explainable Artificial Intelligence (XAI). The proposed approach utilizes the Student Stress Factors dataset (1,100 instances and 20 features) and incorporates feature engineering, feature selection, and dimensionality reduction before evaluating eleven machine learning and ensemble models. Experimental results show that feature engineering combined with normalization significantly improves predictive performance. The best performing model, Random Forest under the C4_FE_Norm configuration, achieved an Overall Score of 92.58%, an Accuracy of 89.55%, and a Brier Score of 0.0396. SHAP provides both global and local explanations, improving model transparency while preserving predictive performance. The final model was implemented in a Streamlit-based platform, and its recommendation templates were verified by psychologists. Usability evaluation achieved an average System Usability Scale (SUS) score of 82.24 (Excellent) and a Single Ease Question (SEQ) score of 6.15/7, indicating that the proposed platform is accurate, interpretable, and user-friendly for early academic burnout risk screening.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Burnout Akademik, Explainable AI, Machine Learning, Platform Berbasis Web, SHAP, Academic Burnout, Explainable Artificial Intelligence, Machine Learning, SHAP, Web-Based Platform. |
| Subjects: | T Technology > T Technology (General) T Technology > T Technology (General) > T385 Visualization--Technique T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.6 Management information systems T Technology > T Technology (General) > T58.62 Decision support systems |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Nabilah Atika Rahma |
| Date Deposited: | 22 Jul 2026 07:33 |
| Last Modified: | 22 Jul 2026 07:33 |
| URI: | http://repository.its.ac.id/id/eprint/136522 |
Available Versions of this Item
- Pengembangan Platform Berbasis Web Untuk Analisis Dan Interpretasi Faktor Risiko Burnout Akademik Menggunakan Machine Learning Dengan Pendekatan Explainable AI. (deposited 22 Jul 2026 07:33) [Currently Displayed]
Actions (login required)
 |
View Item |