Application of Machine Learning to Predict COVID-19 Spread via an Optimized BPSO Model.

Area of Science:

• Epidemiology
• Machine Learning
• Computational Biology

Background:

• Coronavirus disease (COVID-19) case numbers varied significantly across regions.
• Geographical factors like altitude may influence disease spread and prediction accuracy.

Purpose of the Study:

• To develop and evaluate an enhanced predictive model for COVID-19 cases in Saudi Arabia.
• To compare model performance in high-altitude (Taif) versus sea-level (Jeddah) regions.

Main Methods:

• Utilized Binary Particle Swarm Optimization (BPSO) for feature selection.
• Implemented and compared three machine learning models: Random Forest, Gradient Boosting, and Naive Bayes.
• Trained and tested models on datasets from Taif and Jeddah, Saudi Arabia.

Main Results:

• Gradient Boosting achieved 94.6% accuracy in Taif.
• Random Forest achieved 95.5% accuracy in Jeddah.
• The Jeddah dataset yielded better overall accuracy than the Taif dataset.

Conclusions:

• The enhanced model effectively predicts COVID-19 cases in different Saudi Arabian regions.
• Sea-level regions (Jeddah) showed higher prediction accuracy than high-altitude regions (Taif).
• Model performance varied by geographical location and chosen machine learning algorithm.