Forecasting
for Disease
Biosurveillance.
My research focuses on county-level machine learning systems for Highly Pathogenic Avian Influenza (HPAI), with an emphasis on forward-looking forecasting, leakage-safe evaluation, and interpretable risk outputs.
Forecasting Framework
Modeling HPAI outbreak risk across the conterminous United States using environmental, agricultural, climatological, and geospatial data streams.
Building forward-only forecasting infrastructure with time-gated feature construction, persisted artifacts, and monthly backtesting.
Evaluating graph-based and spatiotemporal ML model architectures over county adjacency networks, capturing regional disease dynamics.
Designing evaluation workflows for highly imbalanced data (~1% positives), prioritizing balanced accuracy and recall.
Translating model outputs into interpretable county-level risk maps, diagnostics, and interactive web-based tools for disease biosurveillance.