The XBART-Poisson Classification Model for COVID-19 Data Analysis in Egypt

  • Hanaa Elgohari Department of applied statistics, faculty of commerce, Mansoura University, Egypt
DOI: https://doi.org/10.19139/soic-2310-5070-2738
Keywords: XBART-Poisson, Covid-19, prediction, Mean Absolute Error, Accuracy and ROC-AUC, Precision, Recall, and F1 Score

Abstract

This paper aims to predict daily case, mortality counts, classify high-risk periods and provide interpretable, probabilistic insights into COVID-19 trends in Egypt by using Extreme Bayesian Additive Regression Trees with Poisson likelihood (XBART-Poisson) model to COVID-19 data in Egypt. The model is adapted for the pandemic's count-based data, such as daily cases, mortality counts, and recovery rates, offering a Bayesian probabilistic approach to forecast trends and analyze epidemiological factors. The Poisson likelihood effectively handles the discrete nature of these data points. Performance is benchmarked against traditional classification techniques, revealing XBART-Poisson’s robustness in capturing key trends and providing accurate predictions for COVID-19 progression in Egypt. The study reaches the suggested model which is more accurate than the traditional models such as Logistic Regression, Decision Tree, Random Forest and XGBoost.
Published
2025-11-12
How to Cite
Elgohari, H. (2025). The XBART-Poisson Classification Model for COVID-19 Data Analysis in Egypt. Statistics, Optimization & Information Computing, 14(6), 3761-3775. https://doi.org/10.19139/soic-2310-5070-2738
Issue
Vol 14 No 6 (2025)
Section
Research Articles
Authors who publish with this journal agree to the following terms:

    1. Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
    2. Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
    3. Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).