Development of a Machine Learning Model for Predicting Graduate Learning Outcome Achievement in Outcome-Based Vocational Education

DOI

10.2991/978-94-6239-805-4_9How to use a DOI?

Keywords

Outcome-Based Education; Graduate Learning Outcomes; Machine Learning; Prediction Model; Vocational Education

Abstract

Outcome-Based Education (OBE) requires continuous monitoring of Graduate Learning Outcome (GLO) achievement, yet in most Indonesian vocational institutions this monitoring remains a retrospective, end-of-program exercise rather than a tool for early intervention. This study develops and evaluates a machine learning approach for predicting GLO achievement using institutional academic assessment records of 137 first-year students (2024 cohort) enrolled in the D3 Accounting program at Politeknik Negeri Bali. After excluding 29 students whose GLO indicators had not yet been assessed, 108 records covering 24 measurable indicators were analyzed. Knowledge- and Attitude-domain scores were used as predictors for a composite Skills-domain achievement outcome, modeled both as a continuous score (regression) and as OBE-defined achievement categories (classification); the two framings were combined because a continuous score supports fine-grained risk ranking while a categorical output maps directly onto the discrete achievement bands academic advisors already use. Five regression algorithms and six classification algorithms were compared using 10-fold and leave-one-out cross-validation, respectively. Ridge Regression achieved the best regression performance (R² = 0.97, MAE = 0.32), while Gaussian Naive Bayes achieved the best macro-F1 (0.80) for the imbalanced three-class categorization. Correlation analysis further revealed that several indicators, despite representing conceptually distinct competencies, carried identical scores for nearly all students — a data-integrity artifact traced to the institution’s indicator-to-course mapping rather than genuine pedagogical association. The findings demonstrate the technical feasibility of GLO prediction from partial assessment data while underscoring the need for cleaner, more granular, and longitudinal data before such models can be reliably deployed for early-warning decision-making.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - I Putu Gede Abdi Sudiatmika
AU  - I Made Wijana
AU  - Dewa Ayu Mas Putriari Nusantari
AU  - Ni Ketut Nadila Suryasari
AU  - Putu Satya Saputra
AU  - Ni Made Wirasyanti Dwi Pratiwi
PY  - 2026
DA  - 2026/10/08
TI  - Development of a Machine Learning Model for Predicting Graduate Learning Outcome Achievement in Outcome-Based Vocational Education
BT  - Proceedings of the International Conference on Sustainable Green Tourism Applied Science - Engineering Applied Science 2026 (ICOSTAS-EAS 2026)
PB  - Atlantis Press
SP  - 73
EP  - 84
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6239-805-4_9
DO  - 10.2991/978-94-6239-805-4_9
ID  - Sudiatmika2026
ER  -