EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Interpretable machine learning for health outcomes: SHAP-driven insights from random forest analysis

O.P. Adebayo, I. Ahmed, M.I. Garba, K. T. Oyeleke

Journal of Statistical Sciences and Computational Intelligence · 14.04.2026 · 2 (2) · 253-270

DOI: 10.64497/jssci.24 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Statistical Sciences and Computational Intelligence, online erschienen am 14.04.2026. Verfasst von O.P. Adebayo und 3 weiteren Autor:innen. Die Arbeit nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study develops and compares interpretable machine learning models to predict health insurance utilization patterns using clinical and demographic data. Addressing severe class imbalance (629 "no" vs 8173 "yes" cases) through SMOTE oversampling, we trained and evaluated three distinct models: Random Forest (RF), Logistic Regression, and LASSO regression. The RF model demonstrated superior performance with mean ROC-AUC of 0.834 (95% CI: 0.819-0.859) across 5-fold cross-validation, compared to 0.782 for Logistic and 0.783 for LASSO models. SHAP (SHapley Additive exPlanations) analysis revealed age (β = -0.029, p

Zitieren

APA

Adebayo, O., Ahmed, I., Garba, M., & Oyeleke, K. T. (2026). Interpretable machine learning for health outcomes: SHAP-driven insights from random forest analysis. Journal of Statistical Sciences and Computational Intelligence, 2(2), 253-270. https://doi.org/10.64497/jssci.24

Vancouver

Adebayo O, Ahmed I, Garba M, Oyeleke KT. Interpretable machine learning for health outcomes: SHAP-driven insights from random forest analysis. Journal of Statistical Sciences and Computational Intelligence. 2026;2(2):253-270. doi:10.64497/jssci.24

BibTeX
@article{Adebayo2026,
  author = {Adebayo, O.P. and Ahmed, I. and Garba, M.I. and Oyeleke, K. T.},
  title = {Interpretable machine learning for health outcomes: SHAP-driven insights from random forest analysis},
  journal = {Journal of Statistical Sciences and Computational Intelligence},
  year = {2026},
  volume = {2},
  number = {2},
  pages = {253--270},
  doi = {10.64497/jssci.24}
}
Kostenlos starten →