ANALISIS KARAKTERISTIK AKADEMIK MAHASISWA MENGGUNAKAN K-MEANS CLUSTERING BERDASARKAN TREN NILAI DAN HAMBATAN UJIAN
DOI:
https://doi.org/10.54314/jssr.v9i4.6900Keywords:
k-means, CRISP-DM, akademik, text miningAbstract
Abstract: This study categorizes students from the 2021 cohort based on academic performance dynamics to overcome the limitations of traditional, static GPA-based evaluations. This approach integrates grade growth trend (slope), remedial exam frequency, and thesis topic classification within the CRISP-DM framework. Utilizing the K-Means algorithm, students are segmented into three distinct clusters: High Achiever, At-Risk, and Critical Failure. The empirical results reveal that the group experiencing the sharpest performance decline (slope -0.720) exhibits the highest remedial exam frequency, averaging 3.69 times, and has not yet proposed a research topic. Conversely, students with positive grade trends display mature research readiness, predominantly focusing on Data Mining. These findings confirm a strong correlation between academic schedule interruptions and delayed graduation risks. The resulting taxonomy successfully maps holistic student risk profiles, serving as an effective early warning system for department policymakers to implement personalized and targeted interventions.
Keyword: K-Means; CRISP-DM; academic; text mining.
Abstrak: Penelitian ini mengelompokkan mahasiswa angkatan 2021 berdasarkan dinamika performa akademik guna mengatasi keterbatasan evaluasi tradisional berbasis IPK statis. Pendekatan ini mengintegrasikan variabel tren pertumbuhan nilai (slope), frekuensi ujian susulan, dan klasifikasi topik skripsi dalam kerangka kerja CRISP-DM. Menggunakan algoritma K-Means, mahasiswa dikelompokkan ke dalam tiga klaster: High Achiever, At-Risk, dan Critical Failure. Hasil analisis menunjukkan bahwa kelompok mahasiswa yang mengalami penurunan performa paling tajam (slope -0,720) memiliki rata-rata ujian susulan tertinggi, mencapai 3,69 kali, serta belum mengajukan topik tugas akhir. Sebaliknya, mahasiswa dengan tren nilai positif menunjukkan kesiapan riset yang matang, didominasi oleh topik Data Mining. Temuan ini membuktikan adanya korelasi kuat antara interupsi jadwal akademik dengan risiko keterlambatan kelulusan. Model taksonomi yang dihasilkan sukses memetakan profil risiko mahasiswa secara holistik, sekaligus berfungsi sebagai sistem deteksi dini bagi pengambil kebijakan program studi untuk melakukan intervensi yang lebih personal dan tepat sasaran.
Kata kunci: K-Means; CRISP-DM; akademik; text mining.
Downloads
References
Asif, R. et al. (2017) ‘Analyzing undergraduate students’ performance using educational data mining’, Computers & Education, 113, pp. 177–194. Available at: https://doi.org/https://doi.org/10.1016/j.compedu.2017.05.007.
Danu, R. et al. (2025) ‘Pengelompokkan Data Nilai Mahasiswa Menggunakan Metode K-Means’, Jurnal Algoritme, 5(3), pp. 350–361. Available at: https://doi.org/10.35957/algoritme.v6i1.11313.
Govindasamy, K. and Thambusamy, V. (2018) ‘Analysis of student academic performance using clustering techniques’, International Journal of Pure and Applied Mathematics, 119, pp. 309–322.
Hussain, S. et al. (2018) ‘Educational data mining and analysis of students’ academic performance using WEKA’, Indonesian Journal of Electrical Engineering and Computer Science, 9(2), pp. 447–459. Available at: https://doi.org/10.11591/ijeecs.v9.i2.pp447-459.
M., R., F., N. and A., A. (2018) ‘Predicting and Analysis of Students’ Academic Performance using Data Mining Techniques’, International Journal of Computer Applications, 182(32), pp. 1–6. Available at: https://doi.org/10.5120/ijca2018918250.
Romero, C. and Ventura, S. (2020) ‘Educational data mining and learning analytics: An updated survey’, WIREs Data Mining and Knowledge Discovery, 10(3), p. e1355. Available at: https://doi.org/https://doi.org/10.1002/widm.1355.
Schröer, C., Kruse, F. and Gómez, J.M. (2021) ‘A Systematic Literature Review on Applying CRISP-DM Process Model’, Procedia Computer Science, 181, pp. 526–534. Available https://doi.org/https://doi.org/10.1016/j.procs.2021.01.199.
Shafiq, D.A. et al. (2022) ‘Student Retention Using Educational Data Mining and Predictive Analytics: A Systematic Literature Review’, IEEE Access, 10(July), pp. 72480–72503. Available at: https://doi.org/10.1109/ACCESS.2022.3188767.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Husnul Khair, Tengku Didi Ferdillah (Author)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
