Analisis Komparatif Determinan Prestasi Matematika dan Bahasa Indonesia Menggunakan Random Forest dan SHAP
DOI:
https://doi.org/10.35968/jsi.v13i2.2111Kata Kunci:
Educational Data Mining, Prediksi Kinerja Siswa, Feature Importance, SHAP, Random ForestAbstrak
Penelitian ini bertujuan membandingkan determinan prestasi akademik siswa antara mata pelajaran Matematika dan Bahasa Indonesia menggunakan pendekatan Educational Data Mining. Menggunakan dataset UCI Student Performance yang terdiri dari 395 siswa Matematika dan 649 siswa Bahasa Indonesia, penelitian mengidentifikasi 382 siswa yang terdaftar di kedua mata pelajaran. Metode Random Forest dioptimasi dengan Grid Search dan dievaluasi menggunakan RMSE, MAE, dan R², sedangkan analisis interpretabilitas menggunakan SHAP (SHapley Additive exPlanations). Hasil menunjukkan model lebih akurat pada Bahasa Indonesia (R²=0,85) dibandingkan Matematika (R²=0,78). Analisis feature importance dan SHAP mengungkapkan perbedaan signifikan: prestasi Matematika lebih dipengaruhi oleh ketidakhadiran (r=-0,28), waktu belajar (r=0,31), dan konsumsi alkohol (r=-0,22), sementara prestasi Bahasa Indonesia lebih ditentukan oleh pendidikan ibu (r=0,32) dan ayah (r=0,28). Kesimpulannya, strategi intervensi pendidikan perlu disesuaikan secara spesifik per mata pelajaran.
Referensi
Adawiya, R., & Csíkos, C. (2026). Exploring the interactions between metacognitive strategies, emotional factors, and mathematics performance: Evidence from a CB-SEM model in Indonesian high school students. Acta Psychologica, 262. https://doi.org/10.1016/J.ACTPSY.2025.106131
Awaludin, M., Nuryadi, H., & Pribadi, G. N. (2024). Sistem Otomatisasi Laporan untuk Optimalisasi Pelaporan Data Penelitian dan Pengabdian kepada Masyarakat di Universitas Dirgantara Marsekal Suryadarma. 9675, 1–7.
Banjarnahor, E., Wibawanta, B., Belferik, R., Sadeva, T. A. P., & Andriani, N. (2025). Implementation of Machine Learning Models to Predict First-Year Student Achievement Based on Ethnicity and Entrance Test Scores. 2025 1st International Conference on Data Science and Geoinformatics (ICDSG), 370–375. https://doi.org/10.1109/ICDSG67714.2025.11381374
Cortez, P., & Silva, A. (2008). Using data mining to predict secondary school student performance. 15th European Concurrent Engineering Conference 2008, ECEC 2008 - 5th Future Business Technology Conference, FUBUTEC 2008, 2003(2000), 5–12.
Henry Wasosa. (2025). Influence of Psychological Well-Being and School Factors on Delinquency. International Journal of Research and Innovation in Social Science (IJRISS), VII(2454), 1175–1189. https://doi.org/10.47772/IJRISS
Lailawati, Najmuddin, & Siraj. (2025). Implementation of the Independent Learning Curriculum in Improving Student Learning Outcomes. Jurnal Inovasi Pendidikan Dan Sains, 6(3), 580–586.
MZ, M., Mailizar, M., & Elizar, E. (2025). Understanding Indonesian Students’ Mathematics Performance: A Secondary Analysis of PISA Data. Jurnal Didaktik Matematika, 12(1), 181–195. https://doi.org/10.24815/JDM.V12I1.44700
Rifka Alkhilyatul Ma’rifat, I Made Suraharta, I. I. J. (2024). Independent Curriculum in Improving the Quality of Education Putri. Education Achievment: Journal of Science and Research, 2(2), 306–312.
Satria, D., Permani, R., Winarno, K., Kaluge, D., Indraswari, C. R., & Handrito, R. P. (2025). an Exploratory Study of High-Educated Poverty Through Machine Learning Approach: a Case Study of East Java, Indonesia. Business, Management and Economics Engineering, 23(1), 92–107. https://doi.org/10.3846/bmee.2025.20808
Siti Marhamah Winarti, Nur Ahyani, & Nurlina Nurlina. (2025). The Influence of Independent Curriculum and Classroom Management on Elementary Students’ Learning Outcomes in Ogan Ilir. International Journal of Educational Development, 2(3), 09–18. https://doi.org/10.61132/ijed.v2i3.307
Tiopan Octavianus Sitorus, P. A., Indrianti, Y., Sasmoko, Manalu, S. R., & Anindyo Widhoyoko, S. (2025). Identifying Key Drivers of the Indonesian Entrepreneurial Performance Using Machine Learning and RapidMiner. International Conference on Information Management and Technology, 686–691. https://doi.org/10.1109/ICIMTECH67074.2025.11265240
Unduhan
Diterbitkan
Terbitan
Bagian
Lisensi
Hak Cipta (c) 2026 JSI (Jurnal sistem Informasi) Universitas Suryadarma
Artikel ini berlisensiCreative Commons Attribution-ShareAlike 4.0 International License.
