Peringatan Dini Kegagalan Mata Kuliah Berbasis Kehadiran Awal Semester pada Data SIAKAD
DOI:
https://doi.org/10.70609/jusifor.v5i1.11353Kata Kunci:
peringatan dini, kegagalan mata kuliah, presensi, kebocoran data, regresi logistik FirthAbstrak
Kegagalan mata kuliah adalah tanda awal yang mendahului putus studi. Namun, sebagian besar sistem peringatan dini (early warning system, EWS) di perguruan tinggi Indonesia dibangun berdasarkan nilai akhir semester dan diuji dengan pembagian data acak, sehingga peringatannya datang terlambat dan kinerjanya tampak lebih baik daripada kenyataan. Kami menguji seberapa dini presensi per pertemuan di SIAKAD mampu menandai mahasiswa yang akan gagal. Data berupa 193 pasangan mahasiswa–kelas dari 10 kelas pada tahun akademik 2025/2026 (20 kegagalan, 2.948 catatan presensi). Prediksi dibuat pada pertemuan ke-2, 3, 4, 6, dan 8 hanya dengan informasi yang sudah ada saat itu, memakai aturan ambang sederhana dan regresi logistik Firth, lalu diuji lintas kelas (leave-one-class-out, termasuk varian ketat yang juga memisahkan mahasiswa) dengan interval kepercayaan bootstrap. Absen sekali saja pada dua pertemuan pertama sudah menandai 21,2% mahasiswa dan menangkap 80% kasus gagal; dari mahasiswa yang hadir penuh, hanya 2,6% yang gagal. Model Firth dengan status mengulang mencapai AUC lintas kelas 0,850 [0,721–0,955] pada pertemuan keenam. Sebaliknya, fitur yang baru tersedia setelah cutoff (UAS, total kehadiran) memberi AUC 0,93–0,96, dan kami tunjukkan bahwa angka setinggi itu adalah kebocoran data, bukan kemampuan prediksi.
Referensi
[1] R. Balfanz, L. Herzog, and D. J. Mac Iver, "Preventing student disengagement and keeping students on the graduation path in urban middle-grades schools: Early identification and effective interventions," Educ. Psychol., vol. 42, no. 4, pp. 223–235, 2007, doi: 10.1080/00461520701621079.
[2] M. Credé, S. G. Roch, and U. M. Kieszczynka, "Class attendance in college: A meta-analytic review of the relationship of class attendance with grades and student characteristics," Rev. Educ. Res., vol. 80, no. 2, pp. 272–295, 2010, doi: 10.3102/0034654310362998.
[3] T. Swiderski, S. C. Fuller, and K. C. Bastian, "The relationship between student attendance and achievement, pre- and post-COVID," AERA Open, vol. 11, no. 1, pp. 1–19, 2025, doi: 10.1177/23328584251371041.
[4] L. P. Macfadyen and S. Dawson, "Mining LMS data to develop an 'early warning system' for educators: A proof of concept," Comput. Educ., vol. 54, no. 2, pp. 588–599, 2010, doi: 10.1016/j.compedu.2009.09.008.
[5] K. E. Arnold and M. D. Pistilli, "Course Signals at Purdue: Using learning analytics to increase student success," in Proc. 2nd Int. Conf. Learning Analytics and Knowledge (LAK '12), Vancouver, Canada, 2012, pp. 267–270, doi: 10.1145/2330601.2330666.
[6] S. M. Jayaprakash, E. W. Moody, E. J. M. Lauría, J. R. Regan, and J. D. Baron, "Early alert of academically at-risk students: An open source analytics initiative," J. Learn. Anal., vol. 1, no. 1, pp. 6–47, 2014, doi: 10.18608/jla.2014.11.3.
[7] A. M. Romadhoni, W. P. P. Witra, and I. G. Anugrah, "Penerapan algoritma random forest sebagai early warning system prediksi kelulusan mahasiswa tepat waktu," J. Inf. Technol. Comput. Sci. (INTECOMS), vol. 9, no. 3, pp. 459–470, 2026.
[8] P. Prima, A. R. Adriansyah, and A. N. Usyaid, "Prediksi retensi mahasiswa menggunakan algoritma random forest dengan optimasi algoritma genetika," MIND J., vol. 11, no. 1, pp. 115–126, 2026, doi: 10.26760/mindjournal.v11i1.115-126.
[9] T. A. Marzuqi, E. Kristiani, and Marcel, "Prediksi mahasiswa drop-out di Universitas XYZ," J. Teknol. Inf. Ilmu Komput. (JTIIK), vol. 11, no. 6, pp. 1345–1350, 2024, doi: 10.25126/jtiik.2024118689.
[10] V. Realinho, J. Machado, L. Baptista, and M. V. Martins, "Predicting student dropout and academic success," Data, vol. 7, no. 11, Art. no. 146, 2022, doi: 10.3390/data7110146.
[11] F. A. Putra, S. Mirajdandi, Nandra, B. Okmarizal, and S. Mulyanda, "Prediksi dropout mahasiswa: Early-warning berbasis enrollment dengan machine learning," J. Fasilkom, vol. 15, no. 3, pp. 465–473, 2025, doi: 10.37859/jf.v15i3.10714.
[12] L. Handayani and Priyadi, "A machine learning-based early warning system for student performance prediction: System development and empirical evaluation in higher education," J. Artif. Intell. Inf. Technol. (JAI-IT), vol. 1, no. 1, pp. 171–190, 2026. [Online]. Available: https://journal.perkivi.or.id/index.php/jai-it/article/view/78
[13] T. Fauszt, "Time-consistent prediction in higher education: A framework for preventing data leakage in longitudinal models," Information, vol. 17, no. 6, Art. no. 581, 2026, doi: 10.3390/info17060581.
[14] S. Kaufman, S. Rosset, C. Perlich, and O. Stitelman, "Leakage in data mining: Formulation, detection, and avoidance," ACM Trans. Knowl. Discov. Data, vol. 6, no. 4, pp. 1–21, 2012, doi: 10.1145/2382577.2382579.
[15] A. Shaikhanova, O. Kuznetsov, K. Iklassova, A. Tokkuliyeva, and L. Sugurova, "Interpretable predictive modeling for educational equity: A workload-aware decision support system for early identification of at-risk students," Big Data Cogn. Comput., vol. 9, no. 11, Art. no. 297, 2025, doi: 10.3390/bdcc9110297.
[16] F. Marbouti, H. A. Diefes-Dux, and K. Madhavan, "Models for early prediction of at-risk students in a course using standards-based grading," Comput. Educ., vol. 103, pp. 1–15, 2016, doi: 10.1016/j.compedu.2016.09.005.
[17] H.-C. Chen, E. Prasetyo, S.-S. Tseng, K. T. Putra, Prayitno, S. S. Kusumawardani, and C.-E. Weng, "Week-wise student performance early prediction in virtual learning environment using a deep explainable artificial intelligence," Appl. Sci., vol. 12, no. 4, Art. no. 1885, 2022, doi: 10.3390/app12041885.
[18] D. Firth, "Bias reduction of maximum likelihood estimates," Biometrika, vol. 80, no. 1, pp. 27–38, 1993, doi: 10.1093/biomet/80.1.27.
[19] G. Heinze and M. Schemper, "A solution to the problem of separation in logistic regression," Stat. Med., vol. 21, no. 16, pp. 2409–2419, 2002, doi: 10.1002/sim.1047.
[20] P. Peduzzi, J. Concato, E. Kemper, T. R. Holford, and A. R. Feinstein, "A simulation study of the number of events per variable in logistic regression analysis," J. Clin. Epidemiol., vol. 49, no. 12, pp. 1373–1379, 1996, doi: 10.1016/S0895-4356(96)00236-3.
[21] M. S. Rahman and M. Sultana, "Performance of Firth- and log F-type penalized methods in risk prediction for small or sparse binary data," BMC Med. Res. Methodol., vol. 17, Art. no. 33, 2017, doi: 10.1186/s12874-017-0313-9.
[22] T. Saito and M. Rehmsmeier, "The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets," PLoS ONE, vol. 10, no. 3, p. e0118432, 2015, doi: 10.1371/journal.pone.0118432.
[23] E. W. Steyerberg, A. J. Vickers, N. R. Cook, T. Gerds, M. Gonen, N. Obuchowski, M. J. Pencina, and M. W. Kattan, "Assessing the performance of prediction models: A framework for traditional and novel measures," Epidemiology, vol. 21, no. 1, pp. 128–138, 2010, doi: 10.1097/EDE.0b013e3181c30fb2.
[24] A. Fatunnisa and H. Marcos, "Prediksi kelulusan tepat waktu siswa SMK teknik komputer menggunakan algoritma random forest," J. Manaj. Inform. (JAMIKA), vol. 14, no. 1, pp. 101–111, 2024, doi: 10.34010/jamika.v14i1.12114.
[25] B. Ujkani, D. Minkovska, and N. Hinov, "Course success prediction and early identification of at-risk students using explainable artificial intelligence," Electronics, vol. 13, no. 21, Art. no. 4157, 2024, doi: 10.3390/electronics13214157.
[26] Y. Liu, W. Wang, and E. Xu, "The effectiveness of learning analytics-based interventions in enhancing students' learning effect: A meta-analysis of empirical studies," SAGE Open, vol. 15, no. 2, pp. 1–22, 2025, doi: 10.1177/21582440251336707.
Unduhan
Diterbitkan
Terbitan
Bagian
Lisensi
Hak Cipta (c) 2026 Sani Nurmalasari, Ircham Ali, Verry Verry
Artikel ini berlisensiCreative Commons Attribution-ShareAlike 4.0 International License.
