Application of explainable artificial intelligence approach to predict student learning outcomes

Jean-Baptiste M.B. Sanfo ()
Additional contact information
Jean-Baptiste M.B. Sanfo: The University of Shiga Prefecture

Journal of Computational Social Science, 2025, vol. 8, issue 1, No 9, 33 pages

Abstract: Abstract Machine learning can help predict critical educational outcomes, but its “black-box” nature is a significant challenge for its broad adoption in educational settings. This study employs a variety of supervised learning algorithms applied to data from Burkina Faso’s 2019 Program for the Analysis of CONFEMEN Education Systems. Shapley Additive Explanation (SHAP) is then used on the selected algorithms to identify the most significant factors influencing student learning outcomes. The objectives of the study are to (1) to apply and evaluate supervised learning models (classification and regression) to achieve the highest performance in predicting student learning outcomes; (2) to apply Shapley Additive Explanation (SHAP) to extract the features with the highest predictive power of students’ learning outcomes. Results showed that K-Nearest Neighbors (KNN) and Support Vector Machine (SVM) have the best predictive power for classification tasks. Likewise, the Random Forest Regressor showed the best predictive accuracy for the regression task. SHAP values were then utilized to determine feature contribution to predictions. The key predictive features identified are “local development,” “community involvement,” “school infrastructure,” and “teacher years of experience.” These findings suggest that learning outcomes are significantly influenced by community and infrastructural factors and teacher experience. The implications of this study are substantial for educational policymakers and practitioners. Emphasizing “local development” and “community involvement” underscores the necessity of community engagement programs and partnerships. Prioritizing investments in school infrastructure can enhance the learning environment, while recognizing the impact of teacher years of experience highlights the need for professional development and retention strategies for educators. These insights advocate for a comprehensive approach to improving educational outcomes through targeted investments and strategic community collaborations.

Keywords: Machine learning; Learning outcomes; Supervised learning; XAI; AIED (search for similar items in EconPapers)
Date: 2025
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s42001-024-00344-w Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:jcsosc:v:8:y:2025:i:1:d:10.1007_s42001-024-00344-w

Ordering information: This journal article can be ordered from
http://www.springer. ... iences/journal/42001

DOI: 10.1007/s42001-024-00344-w

Access Statistics for this article

Journal of Computational Social Science is currently edited by Takashi Kamihigashi

More articles in Journal of Computational Social Science from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().