Forecasting of Monkeypox Cases Using Optimized SARIMAX Based Model
Paper ID : 1140-ICEEM2023 (R1)
Authors
Sayed kenawy *1, Mahmoud Elshabrawy2, Marwa M. Eid3, Abdelaziz A. Abdelhamid4, Abdelhameed Ibrahim5
1Department of Communications and Electronics, Delta Higher Institute of Engineering and Technology
2Department of Communications and Electronics Delta Higher Institute of Engineering and Technology, Mansoura, Egypt
3Faculty of Artificial Intelligence Delta University for Science and Technology Mansoura 35111, Egypt
4Department of Computer Science Faculty of Computer and Information Sciences Ain Shams University, Cairo, Egypt
5Computer Engineering and Control Systems Department Faculty of Engineering University Mansoura, Egypt
Abstract
This study presents a dipper-throated-based ant
colony optimization (DTACO) with the Seasonal Auto-Regressive
Integrated Moving Average with eXogenous factor (SARIMAX)
model (DTACO+SARIMAX) to forecast monkeypox cases. The
work optimizes the SARIMAX model using grid search crossvalidation
and fine-tunes its hyperparameters using DTACO to
improve prediction accuracy. The suggested model’s consistency
and accuracy are considerable compared to previous studies.
Comparisons with state-of-the-art models validate the proposed
model’s predictions. DTACO+SARIMAX can be used to control
disease and monitor monkeypox. Healthcare organizations and
governments can better manage and track the pandemic’s course
by offering accurate predictions, reducing public panic, and
enabling effective pandemic planning. The Analysis of Variance
(ANOVA) and Wilcoxon signed-rank tests are conducted on the
proposed DTACO-SARIMAX model and compared models.
Keywords
Dipper throated optimization, SARIMAX model, ant colony optimization, Monkeypox
Status: Accepted