Data-driven analysis and predictive modeling on COVID-19

Area of Science:

• Epidemiology
• Data Science
• Machine Learning

Background:

• The COVID-19 pandemic has caused widespread global impact since 2019.
• Understanding demographic and intervention effects is crucial for pandemic management.

Purpose of the Study:

• To develop a data-driven analytical model for COVID-19.
• To investigate the pandemic's impact on different genders and age groups.
• To assess the effectiveness of safety measures on virus transmission rates.

Main Methods:

• Utilized machine learning and ensemble models for prediction.
• Analyzed three key aspects: patient gender, global growth rate, and social distancing.
• Employed classic classifiers, bagging, feature-based ensembles, voting, and stacking.

Main Results:

• Demonstrated superior prediction performance compared to existing methods.
• Validated models on three extensive public datasets.
• Identified significant patterns in gender-specific impacts and intervention effectiveness.

Conclusions:

• The proposed machine learning model offers robust analytical capabilities for COVID-19.
• The findings provide insights into pandemic dynamics and control strategies.
• Data-driven approaches are effective for understanding and predicting infectious disease spread.