Predicting the growth and trend of COVID-19 pandemic using machine learning and cloud computing

Area of Science:

• Epidemiology
• Data Science
• Public Health

Background:

• The COVID-19 pandemic (caused by SARS-CoV-2) presents a significant global health crisis.
• Rapidly increasing incidence necessitates advanced methods for tracking and prediction.
• Machine Learning (ML) and Cloud Computing offer potential solutions for epidemic management.

Purpose of the Study:

• To apply an improved mathematical model for analyzing and predicting COVID-19 epidemic growth.
• To develop an ML-based prediction framework for assessing global COVID-19 threats.
• To leverage cloud computing for accurate, real-time epidemic behavior prediction.

Main Methods:

• Utilized an improved mathematical model incorporating iterative weighting for Generalized Inverse Weibull distribution fitting.
• Developed and deployed an ML-based prediction framework on a cloud computing platform.
• Employed a data-driven approach for enhanced prediction accuracy.

Main Results:

• Demonstrated that iterative weighting improves the fit of the Generalized Inverse Weibull distribution for epidemic modeling.
• Achieved more accurate and real-time predictions of epidemic growth behavior through the cloud-deployed ML model.
• Validated the utility of a data-driven approach for proactive public health responses.

Conclusions:

• The developed ML-based prediction framework offers a valuable tool for understanding and managing the COVID-19 pandemic.
• Accurate, real-time epidemic prediction is crucial for effective government and citizen responses.
• Identified further research opportunities and practical applications for advanced epidemic modeling.