COVID-19 Prediction and Forecasting Using Machine Learning

The COVID-19 pandemic has underscored the need for accurate predictive tools to manage pub- lic health crises. This study explores the application of supervised machine learning (ML) techniques, specifically Linear Regression (LR) and Support Vector Machines (SVM), to forecast COVID-19 cases. Utilizing a global dataset of daily confirmed, recovered, and death cases from January 2020 onwards, the research preprocesses the data and trains models to predict trends over a 10-day horizon. Results in- dicate that LR provides consistent and reliable forecasts, estimating an average daily increase of 29,900 confirmed cases globally, while SVM struggles with data fluctuations. These findings highlight the po- tential of ML in enhancing pandemic response strategies, offering actionable insights for healthcare authorities. The study concludes that LR is more suitable for short-term COVID-19 forecasting due to its simplicity and interpretability.