Analyzing Predictors of Schools’ Performance using Machine Learning Methods: Case of Primary and Secondary Schools in Slovakia
(2024) DABN01 20241Department of Economics
Department of Statistics
- Abstract
- This thesis explores the determinants of educational performance in Slovak schools using advanced machine-learning (ML) techniques. It identifies key factors influencing academic outcomes and evaluates the effectiveness of various ML models, including Random Forest, Gradient Boosting, and Neural Networks, among others. This study compiled a complex dataset of 1409 primary and 656 secondary schools and matched it to a variety of demographic and economic characteristics. Results indicate that ensemble tree methods, particularly XGBoost, outperform other models in terms of predictive accuracy. These models consistently identify the higher-educated population in the region, the ratio of teachers to students, and the number of pupils in a... (More)
- This thesis explores the determinants of educational performance in Slovak schools using advanced machine-learning (ML) techniques. It identifies key factors influencing academic outcomes and evaluates the effectiveness of various ML models, including Random Forest, Gradient Boosting, and Neural Networks, among others. This study compiled a complex dataset of 1409 primary and 656 secondary schools and matched it to a variety of demographic and economic characteristics. Results indicate that ensemble tree methods, particularly XGBoost, outperform other models in terms of predictive accuracy. These models consistently identify the higher-educated population in the region, the ratio of teachers to students, and the number of pupils in a school as the most significant predictors of academic performance. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/9155049
- author
- Lelkaite, Gabriele LU and Sokoláková, Miriama LU
- supervisor
- organization
- course
- DABN01 20241
- year
- 2024
- type
- H1 - Master's Degree (One Year)
- subject
- keywords
- Machine Learning, Educational Performance, Slovak Schools, Feature Importance
- language
- English
- id
- 9155049
- date added to LUP
- 2024-09-24 08:34:20
- date last changed
- 2024-09-24 08:34:20
@misc{9155049, abstract = {{This thesis explores the determinants of educational performance in Slovak schools using advanced machine-learning (ML) techniques. It identifies key factors influencing academic outcomes and evaluates the effectiveness of various ML models, including Random Forest, Gradient Boosting, and Neural Networks, among others. This study compiled a complex dataset of 1409 primary and 656 secondary schools and matched it to a variety of demographic and economic characteristics. Results indicate that ensemble tree methods, particularly XGBoost, outperform other models in terms of predictive accuracy. These models consistently identify the higher-educated population in the region, the ratio of teachers to students, and the number of pupils in a school as the most significant predictors of academic performance.}}, author = {{Lelkaite, Gabriele and Sokoláková, Miriama}}, language = {{eng}}, note = {{Student Paper}}, title = {{Analyzing Predictors of Schools’ Performance using Machine Learning Methods: Case of Primary and Secondary Schools in Slovakia}}, year = {{2024}}, }