Segmentation of Alumni Employability Profiles Using the K-Means Algorithm Based on Tracer Study Data
Integrated Journal of Artificial Intelligence and Data Science · 16.11.2026 · 2 (2) · 41-52
DOI: 10.68300/ijaids.v2i2.108 ↗Kurz gefasst
Fachartikel in der Zeitschrift Integrated Journal of Artificial Intelligence and Data Science, online erschienen am 16.11.2026. Verfasst von Firman Aziz und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
Alumni employability serves as a key indicator for evaluating the relevance and quality of higher education in relation to the demands of the labor market. While tracer study data is commonly used to gather information on the status of alumni, its utility can be further expanded to identify patterns and characteristics of alumni in greater depth. This study aims to perform an exploratory segmentation of alumni employability profiles using the K-Means algorithm, based on 2024 tracer study data. The dataset comprises 193 alumni and 23 indicators selected to represent characteristics relevant to employability. The research process involved data inspection and transformation, adjustment of scale orientation, Z-score standardization, application of the K-Means algorithm, and evaluation of potential cluster counts using the Silhouette Score, Davies–Bouldin Index (DBI), and Calinski–Harabasz Index (CHI). Testing was conducted for K values ranging from 2 to 6. The evaluation results indicate that K=2 yields the best relative performance among the tested configurations, with a Silhouette Score of 0.2401, a DBI of 1.5042, and a CHI of 73.14. Cluster 0 contains 115 alumni (59.59%), while Cluster 1 contains 78 alumni (40.41%). The Silhouette Score suggests that the separation between groups is limited; therefore, the segmentation results should be interpreted as exploratory patterns rather than a definitive classification of employability. A substantive interpretation of the two clusters is provided based on the indicators and their verified scale orientations. These findings have the potential to support the analysis of alumni characteristics and the planning of competency development programs and career guidance, while accounting for data limitations and the quality of cluster separation.
Zitieren
APA
Aziz, F., Irmawati, I., & Azis, A. L. (2026). Segmentation of Alumni Employability Profiles Using the K-Means Algorithm Based on Tracer Study Data. Integrated Journal of Artificial Intelligence and Data Science, 2(2), 41-52. https://doi.org/10.68300/ijaids.v2i2.108
Vancouver
Aziz F, Irmawati I, Azis AL. Segmentation of Alumni Employability Profiles Using the K-Means Algorithm Based on Tracer Study Data. Integrated Journal of Artificial Intelligence and Data Science. 2026;2(2):41-52. doi:10.68300/ijaids.v2i2.108
BibTeX
@article{Aziz2026,
author = {Aziz, Firman and Irmawati, Irmawati and Azis, Ayu Lestari},
title = {Segmentation of Alumni Employability Profiles Using the K-Means Algorithm Based on Tracer Study Data},
journal = {Integrated Journal of Artificial Intelligence and Data Science},
year = {2026},
volume = {2},
number = {2},
pages = {41--52},
doi = {10.68300/ijaids.v2i2.108}
}
EUVIMED