Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal FaST : Jurnal Sains dan Teknologi
Olivia Kristianti Kusuma
Department of Mathematics, Universitas Pelita Harapan
Computer Science & IT Environmental Science

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Harnessing Predictive Modelling for Education Index : A Dual Approach with Random Forest and Multiple Linear Regression Olivia Kristianti Kusuma; Jessica; Grace Felicia Christy Widjaya; Helena Margaretha; Ferry Vincenttius Ferdinand; Kie Van Ivanky Saputra
FaST - Jurnal Sains dan Teknologi (Journal of Science and Technology) Vol. 10 No. 1 (2026): MAY
Publisher : Universitas Pelita Harapan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.19166/fastjst.v10i1.10367

Abstract

This study explores the predictive modeling of the Education Index (EI) using a dual approach using Random Forest and Multiple Linear Regression (MLR). The data, obtained from "Our World in Data" spanning 1990–2022, integrates socio-economic and infrastructure indicators, including GDP per capita, government spending on education, and access to electricity. This study includes 20 countries that are categorized by income level: Low-Income (Vietnam, Nepal, Myanmar, Pakistan, Zimbabwe), Lower-Middle-Income (Ghana, Bolivia, Cambodia, Egypt, Bangladesh), Upper-Middle-Income (Argentina, Brazil, Peru, Russia, Mexico) and High-Income (Germany, Italy, Portugal, Iceland, Greece). The analysis reveals that Random Forest outperforms MLR in terms of accuracy and lower error rates, while MLR provides better interpretability of variable relationships. With R2 of 99.34% by Random Forest Regression and 94% by Multiple Linear Regression (MLR). Key findings reveal that GDP per capita, primary and secondary completion rates, and internet usage significantly influence EI, underscoring the importance of economic conditions and infrastructure for educational outcomes. This study contributes to the field by offering comparative insights into machine learning and traditional statistical methods for educational analytics, providing a robust basis for policy development to enhance global education standards.
Co-Authors Ferry Vincenttius Ferdinand Grace Felicia Christy Widjaya Helena Margaretha Jessica Kie Van Ivanky Saputra