IDOBE: A New Benchmark for Epidemic Forecasting
Researchers have introduced IDOBE (Infectious Disease Outbreak forecasting Benchmark Ecosystem), a curated collection of epidemiological time series designed to standardize the evaluation of epidemic forecasting methods. The dataset compiles data from multiple repositories spanning over a century of surveillance across U.S. states and global locations. Using derivative-based segmentation, the team generated over 10,000 outbreaks covering outcomes such as cases and hospitalizations for 13 diseases. The work addresses the lack of standardized benchmarks for statistical and machine learning models used in real-time outbreak response, and aims to improve understanding of model performance for novel outbreaks with limited historical data.
Key facts
- IDOBE is a curated collection of epidemiological time series for outbreak forecasting.
- Data spans over a century of surveillance across U.S. states and global locations.
- Derivative-based segmentation generated over 10,000 outbreaks.
- Covers 13 diseases with outcomes including cases and hospitalizations.
- Addresses the lack of standardized benchmark datasets for epidemic forecasting.
- Aims to evaluate statistical and machine learning models for novel outbreaks.
- Published on arXiv with ID 2604.18521.
- The paper is a replace-cross announcement.
Entities
Institutions
- arXiv
Locations
- United States
- U.S. states