Penerapan K-Means Clustering, DBSCAN, dan K-Modes untuk Segmentasi Mahasiswa Baru Berdasarkan Asal Sekolah dan Wilayah Geografis (Studi Kasus: STIKES Guna Bangsa Yogyakarta)
DOI:
https://doi.org/10.58344/locus.v5i8.6017Keywords:
K-Means, DBSCAN, K-Modes, Segmentasi Mahasiswa Baru, Wilayah Geografis, Strategi PromosiAbstract
Perguruan tinggi menghadapi tantangan dalam merumuskan strategi promosi tepat sasaran karena belum memahami pola sebaran mahasiswa baru berdasarkan asal sekolah dan wilayah geografis, menyebabkan alokasi anggaran promosi tidak efisien dan kerja sama dengan sekolah asal tidak terarah. Penelitian ini bertujuan menerapkan K-Means, DBSCAN, dan K-Modes untuk segmentasi mahasiswa baru serta membandingkan kinerja ketiga metode dalam menyelesaikan permasalahan strategi promosi di STIKES Guna Bangsa Yogyakarta. Menggunakan 634 data mahasiswa baru dengan variabel tingkat pendidikan, jurusan, kabupaten, dan provinsi, metode meliputi label encoding dan normalisasi Min-Max untuk K-Means dan DBSCAN, sedangkan K-Modes bekerja langsung pada data kategorikal. Jumlah klaster optimal ditentukan menggunakan metode elbow (k=4), DBSCAN dengan ?=0,45 dan MinPts=8, serta evaluasi menggunakan Silhouette Coefficient dan Davies-Bouldin Index. Hasil menunjukkan K-Means membentuk empat klaster: Nusa Tenggara dengan SMA IPA (29,5%), Jawa dengan SMK kesehatan (24,6%), Jawa dengan SMA IPS (26,5%), dan luar Jawa dengan MA (19,4%). DBSCAN mengidentifikasi 53 mahasiswa (8,36%) sebagai noise, sementara K-Modes menghasilkan centroid berupa kategori aktual dengan performa lebih rendah (Silhouette: 0,193; DBI: 11,978). K-Means unggul secara metrik (Silhouette: 0,412; DBI: 1,024) dan menjawab inefisiensi anggaran dengan panduan prioritas wilayah. K-Modes unggul dalam interpretabilitas centroid untuk kerja sama sekolah terarah. DBSCAN unggul mendeteksi profil unik yang terlewatkan untuk memperluas jangkauan promosi. Ketiga metode bersifat komplementatif dan menjembatani kesenjangan antara data mentah dengan kebutuhan strategi promosi kampus secara operasional.
References
Cahapin, E. L., Malabag, B. A., Santiago Jr, C. S., Reyes, J. L., Legaspi, G. S., & Adrales, K. L. (2023). Clustering of students admission data using k-means, hierarchical, and DBSCAN algorithms. Bulletin of Electrical Engineering and Informatics, 12(6), 3647–3656.
Chaturvedi, A., Green, P. E., & Carroll, J. D. (2001). K-modes clustering. Journal of Classification, 18(1), 35–55.
Flanagan, C. J., & Doyle, W. R. (2024). Measuring geographic opportunity for higher education in US metropolitan areas. Higher Education Policy, 39(1), 149–181.
Fu, Y. C., Fernandez, F., Kao, J. H., & Tseng, K. H. (2022). Does geodemographic segmentation influence higher education opportunity? A spatial investigation of enrollment at one Taiwanese university. Higher Education, 84(5), 1045–1065.
Guanin-Fajardo, J. H., Guaña-Moya, J., & Casillas, J. (2024). Predicting academic success of college students using machine learning techniques. Data, 9(4), 60.
Huang, Z. (1998) . Extensions to the k-means algorithm for clustering large data sets with categorical values. Data Mining and Knowledge Discovery, 2(3), 283–304.
Huang, Z., & Ng, M. K. (1999). A fuzzy k-modes algorithm for clustering categorical data. IEEE Transactions on Fuzzy Systems, 7(4), 446–452.
Ikotun, A. M., Ezugwu, A. E., Abualigah, L., Abuhaija, B., & Heming, J. (2023). K-means clustering algorithms: A comprehensive review, variants analysis, and advances in the era of big data. Information Sciences, 622, 178–210.
Lee, D., & Pirog, M. (2022). Geographical constraints and college decisions: How does for-profit college play in student's choice? Innovative Higher Education, 48(2), 309–328.
Moningkey, M. J. M., Kaparang, D. R., & Sumual, H. (2024). The distribution pattern of new students admissions using the K-Means clustering algorithm. International Journal of Information Technology and Business, 6(2), 1–10.
Pamutha, T., Promthong, W., & Pahlawan, S. (2025). Analyzing and clustering students admission data in Yala Rajabhat University Thailand. Indonesian Journal of Electrical Engineering and Computer Science, 39(2), 1310–1325.
Prastyabudi, W. A., Alifah, A. N., & Nurdin, A. (2024). Segmenting the higher education market: An analysis of admissions data using K-Means clustering. Procedia Computer Science, 234, 96–105.
Rahmawati, D. P., Hidayati, S., Arismawati, P., & Johan, A. W. S. B. (2024). Prospective new college student dashboard: Insights from K-Means clustering with principal component analysis. Inform: Jurnal Ilmiah Bidang Teknologi Informasi dan Komunikasi, 9(2), 137–144.
Rosadi, M. W., Fatonah, N. S., Firmansyah, G., & Akbar, H. (2025). Clustering-based identification of student support needs in higher education transition. JUSIFO (Jurnal Sistem Informasi), 11(2), 133–140.
Valles-Coral, M. A., Salazar-Ramírez, L., Injante, R., Hernandez-Torres, E. A., Juárez-Díaz, J., Navarro-Cabrera, J. R., Pinedo, L., & Vidaurre-Rojas, P. (2022). Density-based unsupervised learning algorithm to categorize college students into dropout risk levels. Data, 7(11), 165.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Muhammad Thoif Junaidi, Bambang Purnomosidi Dwi Putranto
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-ShareAlike 4.0 International (CC-BY-SA). that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
