Predicting students’ completion time based on semesterly performance

Kiki, B, Natalia, F, Ko, CS and Sudirman, S ORCID: 0000-0003-4083-0810 (2025) Predicting students’ completion time based on semesterly performance. ICIC Express Letters Part B Applications, 16 (12). pp. 1303-1310. ISSN 2185-2766

Preview

Text PREDICTING STUDENTS COMPLETION TIME BASED ON SEMESTERLY PERFORMANCE.pdf - Published Version Download (264kB) | Preview

Abstract

Recent years have seen a notable rise in higher education institutions, spurr-ed by competitive pressures among universities. A critical measure of educational quality lies in students’ completion rates, encompassing the duration of their studies as a signif-icant component. Protracted or failed completions not only squander time but also drain financial and energy resources for individuals pursuing higher education and at the same time they can also tarnish an institution’s reputation and accreditation status. In this study, we construct a predictive model for students’ completion time to improve students’ awareness and furnish academic advisors and staff with valuable insights to tailor academic guidance. Employing three classification algorithms – Naïve Bayes, Support Vector Machine, and Neural Network – the study optimizes these algorithms using the Particle Swarm Optimization technique to yield superior results. The Neural Network algorithm emerges as the optimal performer post-optimization, boasting accuracy, precision, speci-ficity, sensitivity, and F1-score metrics of 0.86, 0.86, 0.86, 0.98, and 0.86, respectively, alongside superior confusion matrix outcomes. The model’s efficacy will be harnessed to forecast data for active-status students and disseminated via a dedicated website.

Item Type:	Article
Subjects:	L Education > L Education (General) L Education > LB Theory and practice of education > LB2300 Higher Education Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions:	Computer Science and Mathematics
Publisher:	ICIC International
Date of acceptance:	1 April 2025
Date of first compliant Open Access:	28 September 2026
Date Deposited:	28 Sep 2026 15:16
Last Modified:	28 Sep 2026 15:16
DOI or ID number:	10.24507/icicelb.16.12.1303
URI:	https://researchonline.ljmu.ac.uk/id/eprint/29550

View Item