Comparative Analysis of Optimization Algorithms for Enhancing Artificial Neural Network Accuracy in Predicting Global Monkeypox Cases Trends

Monkeypox (MPXV) was declared a Public Health Emergency of International Concern (PHEIC) by the World Health Organization due to its rapid global spread and public health impact.This study proposes a trend-aware prediction model using an Artificial Neural Network (ANN) evaluated under four schemes: baseline ANN and ANN optimized using Particle Swarm Optimization (PSO), Harris Hawks Optimization (HHO), and Genetic Algorithm (GA).A global cumulative MPXV dataset obtained from Our World in Data, covering the period from June 2022 to June 2024, was utilized.Model performance was evaluated using Root Mean Squared Error (RMSE) and R-squared (R²).Experimental results indicate that the ANN-GA model achieved the best performance, reducing RMSE by 29.59% from 0.196 to 0.138 and improving R² from 0.9936 to 0.9968.These findings demonstrate that metaheuristic optimization, particularly GA, can effectively enhance ANN performance for trend-based prediction of global monkeypox cases and provide a reliable framework to support public health decision-making.