Evaluating Post-Training Employment Outcomes for Workforce Policy through Clustering: A Comparative Study of K-Means, Hierarchical Clustering, Gaussian Mixture Model, and Fuzzy C-Means

Mayrisa Andriyani; Siti Nurwilda; Wahyu Ningtiyas Mergianti; Nurissaidah Ulinnuha

doi:10.30871/jaic.v10i4.13482

Authors

Mayrisa Andriyani UIN Sunan Ampel Surabaya
Siti Nurwilda UIN Sunan Ampel Surabaya
Wahyu Ningtiyas Mergianti UIN Sunan Ampel Surabaya
Nurissaidah Ulinnuha UIN Sunan Ampel Surabaya

DOI:

https://doi.org/10.30871/jaic.v10i4.13482

Keywords:

Clustering, Davies-Bouldin Index, Fuzzy C-means, Post-Training Evaluation, Workforce Policy, Silhouette Score

Abstract

Evaluating the effectiveness of government training programs requires more than a binary employed/unemployed indicator, since participants who find work still differ substantially in salary level, waiting time to employment, job position, gender, and employment sector. Prior studies applying clustering to workforce or socio-economic data have generally compared only two or three algorithms at a time, on datasets from other domains such as banking or health, leaving it unclear which method best segments multidimensional post-training outcome data with mixed numerical and categorical attributes and no natural hard boundaries between groups. This study addresses that gap by comparing four clustering algorithms, K-Means, Hierarchical Clustering, Gaussian Mixture Model, and Fuzzy C-Means, to segment post-training participant data from the Surabaya City Government, with the aim of providing local policymakers with an evidence-based grouping of participants that can inform which sectors and job levels most need follow-up support. The dataset consists of 509 observations with attributes for job position, monthly salary, gender, employment sector, and waiting time. Prior to clustering, the data were preprocessed through cleaning of inconsistent categorical entries, median imputation of missing numerical values, One-Hot Encoding of categorical attributes, and Min-Max normalization of numerical attributes to a common 0-1 scale. Clustering performance was evaluated using Silhouette Score and Davies-Bouldin Index (DBI) across cluster counts k = 2-5. Fuzzy C-Means with five clusters achieved the best overall performance, with a Silhouette Score of 0.649 and a DBI of 0.640; its closest competitor, Ward-linkage Hierarchical Clustering, achieved a comparable Silhouette Score of 0.647 but a markedly higher DBI of 0.872, indicating that FCM produced more compact, well-separated clusters overall even though the two methods separated participants almost equally well. The resulting five clusters show clear policy-relevant differences: the largest cluster, dominated by the industrial sector, has the highest average salary but also the longest waiting time to employment, suggesting that industrial-sector training would benefit from faster competency certification and job-matching support, while smaller, female-dominated clusters in the social sector show shorter waiting times but lower salaries, pointing to a need for upskilling pathways into higher-paying roles. These findings illustrate how clustering can move post-training evaluation beyond simple placement rates toward data-driven, sector-specific recommendations for workforce training policy.

Downloads

Download data is not yet available.

References

[1] S. G. Pane, W. Pramudya, R. C. Amalya, S. N. Aulia, dan P. nabila Pebriani, “Analisis Pengaruh Tenaga Kerja dan Pengangguran Terhadap Pertumbuhan Ekonomi di Indonesia,” Econ. Rev. J., vol. 3, hal. 1204–1214, 2024, doi: 10,56709/mrj.v3i4,401.

[2] P. Sari, Efan, dan R. Syahri, “Algoritma K-Means Clustering: Sebuah Studi Literatur,” J. Inform, hal. 1–7, 2023, doi: 10,12345/juri.

[3] R. Rahmati, A. W. Wijayanto, P. Studi, K. Statistik, dan P. Sains, “Analisis Cluster dengan Algoritma K-Means, Fuzzy C-means dan Hierarchical Clustering (Studi Kasus: Indeks Pembangunan Manusia Tahun 2019),” J. Inform. dan Komput., vol. 5, no. 2, hal. 73–80, 2021, doi: https://dx.doi.org/10,26798/jiko.v5i2,422.

[4] H. Yang, C. Wang, H. Zhang, Y. Zhou, dan B. Luo, “Recognition of Maize Seed Varieties based on Hyperspectral Imaging Technology and Integrated Learning Algorithms,” PeerJ Comput. Sci., hal. 1–20, 2023, doi: 10,7717/peerj-cs,1354.

[5] H. Huang, Z. Liao, X. Wei, dan Y. Zhou, “Combined Gaussian Mixture Model and Pathfinder Algorithm,” entropy, hal. 1–20, 2023, doi: https://doi.org/10,3390/e25060946.

[6] H. Firdaus dan A. Sofro, “Analisa Cluster menggunakan K-Means dan Fuzzy C-Means dalam Pengelompokan Provinsi Menurut Data Intesitas Bencana Alam di Indonesia Tahun 2017-2021,” J. Ilm. Mat., vol. 10, no. 01, hal. 50–60, 2022, doi: https://doi.org/10,26740/mathunesa.v10n1.p50-60.

[7] E. Y. Ahmadov, “Comparative Analysis of K-Means and Fuzzy C-Means Algorithms on Demographic Data using the PCA Method,” Probl. Inf. Technol., vol. 14, no. 1, hal. 15–22, 2023, doi: http://doi.org/10,25045/jpit.v14.i1,03.

[8] A. A. Wani, “Comprehensive Analysis of Clustering Algorithms : Exploring Limitations and Innovative Solutions,” PeerJ Comput. Sci., 2024, doi: 10,7717/peerj-cs,2286.

[9] E. K. A. Mala, S. Rochman, dan H. Suprajitno, “Comparison of Clustering in Tuberculosis using Fuzzy C-Means and K-Means Methods,” Commun. Math. Biol. Neurosci., hal. 1–20, 2022, doi: https://doi.org/10,28919/cmbn/7335.

[10] D. Chicco, A. Campagner, A. Spagnolo, D. Ciucci, dan G. Jurman, “The Silhouette Coefficient and the Davies-Bouldin Index are more Informative than Dunn index, Calinski-Harabasz index, Shannon entropy, and Gap Statistic for Unsupervised Clustering Internal Evaluation of Two Convex Clusters,” 2025, doi: 10,7717/peerj-cs,3309.

[11] M. K. S. Md Amran Hossen Pabel, Biswanath Bhattacharjee, Sonjoy Kumar Dey, Sakib Salam Jamee, Md Omar Obaid, Md Sakib Mia, Sajidul Islam Khan, “Business Analytics for Customer Segmentation: A Comparative Study of Machine Learning Algorithms in Personalized Banking Services,” Iternational J. Econ. Financ. Manag. Sci., hal. 1–13, doi: https://doi.org/10,55640/ijefms/Volume10Issue03-01.

[12] Z. Parinzka, S. Surono, dan A. Thobirin, “Algoritma Support Vector Regression dan Analisis Long Short-Term Memory sebagai Penanganan Missing data,” J. Teknol. Inf. dan Ilmu Komput., vol. 13, no. 1, hal. 73–82, 2026, doi: https://doi.org/10,25126/jtiik,2026131.

[13] Fahrillah dan Z. Fatah, “Pengelompokkan Data Nilai Siswa Madrasah Ta’hiliyah menggunakan Metode K-means Clustering,” J. Ris. Sist. Inf., vol. 2, no. 1, hal. 53–59, 2025, doi: https://doi.org/10,69714/0v1pkz05.

[14] A. Zhu, Z. Hua, Y. Shi, dan Y. Tang, “An Improved K-Means Algorithm Based on Evidence Distance,” entropy, 2021, doi: https://doi.org/10,3390/ e23111550.

[15] M. Sania Fitri Octavia, “Penerapan K-Means dan Fuzzy C-Means untuk Pengelompokan Data Kasus Covid-19 di Kabupaten Indragiri Hilir,” Build. Informatics, Technol. Sci., vol. 3, no. 2, hal. 88–94, 2021, doi: 10,47065/bits.v3i2,1005.

[16] F. A. Totti dan N. Setiyawati, “Perbandingan Algoritma Clustering K-Means, Gaussian Mixture Model, dan Spectral Clustering untuk Facial Emotion Recognition A Comparative Study of K-Means, Gaussian Mixture Model, and,” J. Sist. Inf., vol. 14, hal. 3007–3019, 2025, doi: https://doi.org/10,32520/stmsi.v14i6,5668.

[17] S. H. Marwoto, “A Comparative Analysis of DBSCAN and Gaussian Mixture Model for Clustering Indonesian Provinces Based on Socioeconomic Welfare Indicators,” J. Ilmu Mat. dan Terap., vol. 19, no. 3, hal. 2039–2056, 2025, doi: https://doi.org/10,30598/barekengvol19iss3pp2039-2056.

[18] V. C. Pezoulas et al., “Bayesian Inference-Based Gaussian Mixture Estimation Towards Large-Scale Synthetic Data Generation for In Silico Clinical Trials,” Eng. Med. Biol., vol. 3, hal. 108–114, 2022, doi: https://doi.org/10,1109/OJEMB,2022,3181796.

[19] D. I. Yulianti, T. I. Hermanto, dan M. Defriani, “Analisis Clustering Donor Darah dengan Metode Agglomerative Hierarchical Clustering,” J. Rekayasa Tek. Inform. dan Inf., vol. 3, no. 6, hal. 303–308, 2023, doi: https://djournals.com/resolusi 2111.

[20] S. S. Brigitta Melati Kumarahadi, Hasih Pratiwi, “Penerapan Metode Hierarchical Clustering untuk Pengelompokan Kota/Kabupaten di Indonesia berdasarkan Indikator Kemiskinan,” vol. 11, no. 2, 2023, doi: https://doi.org/10,30646/tikomsin.v11i2,754.

[21] Z. Alamtaha, I. Djakaria, dan N. I. Yahya, “Implementasi Algoritma Hierarchical Clustering dan Non-Hierarchical Clustering untuk Pengelompokkan Pengguna Media Sosial,” J. Stat. an Its Appl., vol. 4, no. 1, hal. 33–43, 2023, doi: 10,20956/ejsa.vi,24830.

[22] F. Rasidia, R. Goejantoro, M. Fathurahman, dan L. S. Komputasi, “Analisis Klaster Menggunakan Metode Average Linkage dengan Validasi Multiscale Bootstrap (Studi Kasus: Indikator Pendidikan di Indonesia Tahun 2021),” J. Eksponensial, vol. 16, no. April, hal. 23–31, 2025, doi: 10,30872/eksponensial.v16i1,1392.

[23] R. Gustriansyah, J. Alie, dan N. Suhandi, “Hierarchical Clustering for Crime Rate Mapping in Indonesia,” J. Ilm., vol. 14, no. 3, hal. 275–283, 2022.

[24] A. Septianingsih, “Pemetaan Kabupaten Kota di Provinsi Jawa Timur berdasarkan Tingkat Kasus Penyakit menggunakan Pendekatan Agglomeratif Hierarchical Clustering,” J. Ilm. Pendidik. Mat. Mat. dan Stat., vol. 3, no. 2, 2022.

[25] A. M. Sarusu, D. Akmila, M. Wijana, dan M. E. Habiby, “Sistem Informasi Manajemen Data Penduduk Berbasis Website,” Intern. Inf. Syst. J., vol. 6, no. 2, hal. 127–136, 2024, doi: https://doi.org/10,32627/internal.v6i2,859.

[26] H. Ramadhan, M. Rizal, A. Kamaludin, M. A. Nasrullah, dan D. Rolliawati, “Comparison of Hierarchical, K-Means and DBSCAN Clustering Methods for Credit Card Customer Segmentation Analysis Based on Expenditure Level,” J. Appl. Informatics Comput., vol. 7, no. 2, hal. 246–251, 2023, doi: https://doi.org/10,30871/jaic.v7i2,5790.

[27] D. A. Tarigan, “Optimization of the K-Means Clustering Algorithm Using Davies Bouldin Index in Iris Data Classification,” vol. 4, no. 1, hal. 545–552, 2023, doi: 10,30865/klik.v4i1,964.

[28] Y. Lin dan S. Chen, "A Centroid Auto-Fused Hierarchical Fuzzy c-Means Clustering," IEEE Trans. Fuzzy Syst., vol. 29, no. 7, hal. 2006-2017, 2021, doi: 10.1109/TFUZZ.2020.2988848.

[29] M. S. Hasibuan, R. Rizal Nul Fikri, dan D. A. Dewi, "Job Clustering Based on AI Adoption and Automation Risk Levels: An Analysis Using the K-Means Algorithm in the Technology and Entertainment Industries," Int. J. Appl. Inf. Manag., vol. 4, no. 2, hal. 54-69, 2024, doi: 10.47738/ijaim.v4i2.83.