Ensemble Machine Learning for Predicting Academic Performance in STEM Education Yields New Insights

Strategic Insights -

A new research study by the Department of Computer Science presents fresh data on artificial intelligence, with a focus on developing a predictive model for STEM students using advanced ensemble machine learning algorithms. The study, which gathered secondary data from three Ethiopian universities, employed techniques such as Random Forest, CatBoost, and Gradient Boosting to assess the academic success of STEM students. The research found that key factors influencing academic success include student transcripts, departmental affiliation, socio-economic status, parental education and occupation, and entrance exam results. Notably, the Gradient Boosting algorithm achieved impressive results, with 93.35% accuracy, 92.69% precision, 93.14% recall, and an F1 score of 92.90%.

Key Takeaways:

  • The study developed a predictive model for STEM students using ensemble machine learning algorithms, with a focus on assessing academic success.
  • The model employed techniques such as Random Forest, CatBoost, and Gradient Boosting to assess the academic success of STEM students.
  • Key factors influencing academic success include student transcripts, departmental affiliation, socio-economic status, parental education and occupation, and entrance exam results.
  • The Gradient Boosting algorithm achieved impressive results, with 93.35% accuracy, 92.69% precision, 93.14% recall, and an F1 score of 92.90%.
  • The study concluded that the findings are crucial for policymakers aiming to enhance the STEM education landscape in Ethiopia.
  • The research encourages future researchers to explore deep learning techniques alongside additional variables for even greater insights.

Statistics:

  • 93.35% accuracy achieved by the Gradient Boosting algorithm
  • 92.69% precision achieved by the Gradient Boosting algorithm
  • 93.14% recall achieved by the Gradient Boosting algorithm
  • 92.90% F1 score achieved by the Gradient Boosting algorithm

Sources:

  • Ensemble machine learning for predicting academic performance in STEM education. Discover Education, 2025,4(1):1-28. The publisher for Discover Education is Springer. A free version of this journal article is available at https://doi-org.sdpl.idm.oclc.org/10.1007/s44217-025-00710-4.
  • NewsRx. Department of Computer Science Researcher Yields New Findings on Machine Learning (Ensemble machine learning for predicting academic performance in STEM education). Education Letter. September 3, 2025; p 78.