ANALISIS TRACER STUDY MENGGUNAKAN K-MEANS CLUSTERING UNTUK EVALUASI KURIKULUM
DOI:
https://doi.org/10.54314/v69fs464Kata Kunci:
K-Means Clustering, Tracer Study, Evaluasi Kurikulum, Data MiningAbstrak
Abstract: Tracer study is an important instrument for evaluating graduate quality and curriculum relevance to labor market needs. This study aims to analyze tracer study data using the K-Means Clustering method to group alumni characteristics and support curriculum evaluation. The dataset consists of 100 records with a relatively balanced composition, including variables such as waiting time for employment, job-field relevance, initial salary, and graduate competencies. The analysis process includes preprocessing, clustering, and evaluation using Silhouette Score, Davies-Bouldin Index, and Calinski-Harabasz Index. The results show that the data are grouped into three clusters representing different levels of job readiness. A Silhouette Score of 0.17, Davies-Bouldin Index of 1.736, and Calinski-Harabasz Index of 23.5 indicate low to moderate clustering quality due to data heterogeneity and overlap between clusters. The findings highlight the need for curriculum improvement, particularly in enhancing technical and soft skills. Future research is recommended to use larger and more balanced datasets, include more diverse variables, and compare clustering methods to improve analysis quality.
Keywords: K-Means Clustering, Tracer Study, Curriculum Evaluation, Data Mining.
Abstrak: Tracer study merupakan instrumen penting dalam mengevaluasi kualitas lulusan dan relevansi kurikulum terhadap kebutuhan dunia kerja. Penelitian ini bertujuan untuk menganalisis data tracer study menggunakan metode K-Means Clustering untuk mengelompokkan karakteristik alumni serta mendukung evaluasi kurikulum. Data yang digunakan sebanyak 100 data dengan komposisi relatif seimbang, meliputi variabel waktu tunggu kerja, kesesuaian bidang kerja, pendapatan awal, serta kompetensi lulusan. Proses analisis meliputi preprocessing, clustering, dan evaluasi menggunakan Silhouette Score, Davies-Bouldin Index, dan Calinski-Harabasz Index. Hasil penelitian menunjukkan bahwa data terbagi menjadi tiga cluster dengan tingkat kesiapan kerja yang berbeda. Nilai Silhouette Score sebesar 0,17, Davies-Bouldin Index sebesar 1,736, dan Calinski-Harabasz Index sebesar 23,5 menunjukkan kualitas clustering rendah hingga sedang, yang disebabkan oleh sifat data yang heterogen dan adanya overlap antar cluster. Hasil ini mengindikasikan perlunya perbaikan kurikulum, khususnya pada peningkatan kompetensi teknis dan soft skills. Penelitian selanjutnya disarankan menggunakan data yang lebih banyak dan lebih seimbang, menambahkan variabel yang lebih beragam, serta membandingkan metode clustering untuk meningkatkan kualitas analisis.
Kata Kunci: K-Means Clustering, Tracer Study, Evaluasi Kurikulum, Data Mining
Unduhan
Referensi
Afriyadi, M. R., Putra, F., & Purnomo, W. A. (2025). Web-based tracer study system design for the Faculty of Computer Science, University of Dharmas Indonesia. Jurnal Ilmiah Sistem Informasi (JUSI), 4(2), 308–319. https://doi.org/10.51903/phmead65
Febrianti, S., Munawir, M., & Fitria, L. (2021). Penerapan metode K-means clustering terhadap alumni berdasarkan kuesioner tracer study. Journal of Informatics and Computer Science, 7(2), 117–122.
Hamid, M. A., Aribowo, D., & Anggraini, R. (2021). Design and development of alumni career information system using PHP MySQL. ELINVO (Electronics, Informatics, and Vocational Education), 6(1),81–89. https://doi.org/10.21831/elinvo.v6i1.30200
Jannah, W. (2025). Sistem informasi tracer study alumni sekolah menengah kejuruan (SMK) Mandiri. ADA Journal of Information System Research, 2(3), 111–120. https://doi.org/10.64366/adajisr.v2i3.91
Lutfi, N. L., Lutfi, A., & Baijuri, A. (2024). Sistem informasi penelusuran alumni (tracer study) berbasis web pada Fakultas Sains dan Teknologi Universitas Ibrahimy. Seminastika, 5(1), 61–70. https://doi.org/10.47002/seminastika.v5i1.793
Mashuri, M., & Assegaff, S. (2023). Sistem informasi alumni (tracer study) berbasis web pada SMA Negeri 2 Bungo. Manajemen Sistem Informasi, 8(3), 449–456.
Maulana, R., & Fathoni, A. (2025). Analisis clustering rekomendasi mata kuliah peminatan berdasarkan karir alumni menggunakan K-means. JOISIE (Journal of Information Systems and Informatics Engineering), 9(1), 109–122.https://doi.org/10.35145/joisie.v9i1.4954
Putra, D. M., & Elmunsyah, H. (2025). Design of an alumni tracer study information system for vocational high schools. Letters in Information Technology Education (LITE), 8(2), 80–89.
Ramadhan, M. I., Nazir, A., Irsyad, M., Sanjaya, S., & Syafria, F. (2026). Clustering analysis using the K-means method to identify alumni satisfaction pattern. MALCOM: Indonesian Journal of Machine Learning and Computer Science, 6(1), 31–41. https://doi.org/10.57152/malcom.v6i1.2401
Septiana, Y., Fitriani, L., Hawariyan, F., Kurniawati, R., & Ulfa, R. L. (2023). Rancang bangun sistem informasi tracer study alumni berbasis website. Jurnal Algoritma, 20(1), 11–21.
Sipayung, A., Utari, C. T., & Sabila, P. C. (2025). Sistem informasi tracer study berbasis web bagi alumni Fakultas Sains dan Teknologi. Jurnal Persegi Bulat, 3(2), 36–45.
