Naive Bayes-Based Prediction Model for University Student Admissions
Journal of Informatics, Game Engineering, and Networking · 15.07.2026 · 1 (2)
DOI: 10.55732/5whemk80 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Informatics, Game Engineering, and Networking, online erschienen am 15.07.2026. Verfasst von Edi Dwi Prasetyo.
Abstract (Originalsprache)
New student admission is a strategic process for higher education institutions because it directly affects academic input quality and the sustainability of the institution. This study develops a prediction model for new-student admission status (accepted/rejected) using the Gaussian Naive Bayes algorithm, applies systematic data preprocessing, and evaluates the resulting model using Accuracy, Precision, Recall, F1-score, ROC-AUC, and the Confusion Matrix. The dataset consists of 2,000 historical admission records with 14 attributes, split into training and testing sets with an 80:20 ratio (1,600 training and 400 testing records). The Naive Bayes model achieved an accuracy of 93.25%, a macro-average precision of 0.925, recall of 0.933, F1-score of 0.929, and a ROC-AUC of 0.987, exceeding the 80% minimum accuracy target. Benchmarking against four other classifiers showed that Decision Tree (99.00%) and Random Forest (98.50%) achieved higher accuracy, Support Vector Machine performed comparably (93.25%), while k-Nearest Neighbor lagged considerably (64.50%) due to its sensitivity to categorical label encoding. Entrance test score and average report-card score were identified as the most influential attributes for the admission decision. This study contributes a comprehensively validated prediction model together with implementation recommendations for the UNUSIDA admission information system.
Zitieren
APA
Prasetyo, E. D. (2026). Naive Bayes-Based Prediction Model for University Student Admissions. Journal of Informatics, Game Engineering, and Networking, 1(2). https://doi.org/10.55732/5whemk80
Vancouver
Prasetyo ED. Naive Bayes-Based Prediction Model for University Student Admissions. Journal of Informatics, Game Engineering, and Networking. 2026;1(2). doi:10.55732/5whemk80
BibTeX
@article{Prasetyo2026,
author = {Prasetyo, Edi Dwi},
title = {Naive Bayes-Based Prediction Model for University Student Admissions},
journal = {Journal of Informatics, Game Engineering, and Networking},
year = {2026},
volume = {1},
number = {2},
doi = {10.55732/5whemk80}
}
EUVIMED