Evaluating K-Means and K-Medoids Using Silhouette Score for Eysenck Personality-Based Clustering of Prospective Students
Journal of Applied Informatics and Computing · 23.04.2026 · 10 (2) · 1818-1827
DOI: 10.30871/jaic.v10i2.12338 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Applied Informatics and Computing, online erschienen am 23.04.2026. Verfasst von Muhammad Arifin und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).
Abstract (Originalsprache)
The selection of an academic major aligned with students’ personality characteristics plays a crucial role in enhancing academic performance and long-term career development. This study compares the performance of the K-Means and K-Medoids algorithms in clustering 118 students from the Bojonegoro region based on introversion–extraversion tendencies, operationalized using Eysenck’s personality framework. Data were collected through a 30-item personality questionnaire measured on a five-point Likert scale and validated through expert-based content validity assessment. Preprocessing involved transforming responses into numerical values to enable distance-based clustering analysis. Four clustering scenarios were evaluated by combining two algorithms (K-Means and K-Medoids) with two distance metrics (Euclidean and Manhattan). Cluster quality was assessed using the Silhouette Score as an internal validation measure. The results show that K-Means with Euclidean Distance achieved the highest Silhouette Score of 0.44013, indicating a moderate cluster structure and outperforming the other configurations. External validation using expert-defined ground truth revealed 100% agreement for the introverted cluster, while the extraverted cluster demonstrated several mismatches, reflecting the heterogeneous and continuous nature of extraversion traits. These findings highlight the importance of algorithm and distance metric selection in personality-based clustering and demonstrate the potential of integrating psychological theory with data mining techniques to support objective and evidence-based academic major recommendation systems.
Zitieren
APA
Arifin, M., Prastya, I. W. D., & Budiani, J. R. (2026). Evaluating K-Means and K-Medoids Using Silhouette Score for Eysenck Personality-Based Clustering of Prospective Students. Journal of Applied Informatics and Computing, 10(2), 1818-1827. https://doi.org/10.30871/jaic.v10i2.12338
Vancouver
Arifin M, Prastya IWD, Budiani JR. Evaluating K-Means and K-Medoids Using Silhouette Score for Eysenck Personality-Based Clustering of Prospective Students. Journal of Applied Informatics and Computing. 2026;10(2):1818-1827. doi:10.30871/jaic.v10i2.12338
BibTeX
@article{Arifin2026,
author = {Arifin, Muhammad and Prastya, Ifnu Wisma Dwi and Budiani, Jauhara Rana},
title = {Evaluating K-Means and K-Medoids Using Silhouette Score for Eysenck Personality-Based Clustering of Prospective Students},
journal = {Journal of Applied Informatics and Computing},
year = {2026},
volume = {10},
number = {2},
pages = {1818--1827},
doi = {10.30871/jaic.v10i2.12338}
}
EUVIMED