COWAVE: A labelled COVID-19 wave dataset for building predictive models

Area of Science:

• Epidemiology
• Data Science
• Public Health

Background:

• The COVID-19 pandemic has strained global healthcare systems, necessitating robust data analysis.
• Extensive COVID-19 data is publicly available, offering opportunities for epidemiological research.
• Understanding and predicting pandemic waves is crucial for resource management.

Purpose of the Study:

• To collate global COVID-19 case data from the World Health Organization (WHO).
• To develop a labeled dataset for supervised learning by defining pandemic waves.
• To establish a benchmark for predictive models and demonstrate dataset utility for wave forecasting.

Main Methods:

• Collected COVID-19 case data from the WHO website.
• Defined multiple criteria for identifying pandemic waves to create data labels.
• Utilized an eXtreme Gradient Boosting (XGBoost) model for baseline performance evaluation.

Main Results:

• A novel, labeled dataset of global COVID-19 waves was created.
• The dataset was demonstrated to be effective for training supervised learning classifiers.
• A baseline performance standard was established for future predictive models.

Conclusions:

• The curated dataset is a valuable resource for epidemiologists and researchers.
• Early prediction of future COVID-19 waves is facilitated by this dataset.
• The dataset supports the development of advanced machine learning models for pandemic surveillance.