Forecasting COVID-19 Infections with Supervised Learning Methods
International Research Journal of Modernization in Engineering Technology and Science · 2025
The COVID-19 pandemic has posed unprecedented challenges to public health systems worldwide, emphasizing the need for accurate forecasting models to inform timely interventions.This mini project explores the application of supervised machine learning methods for predicting the spread of COVID-19 infections.By leveraging historical infection data and relevant epidemiological features, various regression models such as Linear Regression, Support Vector Regression (SVR), and Random Forest Regressor were implemented to forecast future infection trends.The dataset, sourced from publicly available repositories like Johns Hopkins University or WHO, was preprocessed to handle missing values, normalize features, and construct lag variables for temporal dependency.The models were evaluated using performance metrics such as Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and R² score to compare their predictive accuracy.Results indicate that ensemble methods like Random Forest outperformed linear models in capturing non-linear infection patterns.The findings demonstrate the potential of supervised learning in aiding pandemic response through datadriven forecasting, while also highlighting the importance of feature selection, data quality, and model interpretability in health-related AI applications.