Applying Machine Learning Models to Predict West Nile Virus Outbreaks in California and the U.S.

This study applies machine learning to predict West Nile Virus (WNV) outbreaks in California and the U.S., integrating environmental, demographic, land cover, and surveillance data. Analysis reveals that temperature, wind, and average daylight hours are important factors influencing WNV transmission. Comparing national and county-level datasets, we find that higher-resolution, county-level data improves predictive insights by capturing finer spatial and temporal patterns. These findings emphasize the importance of detailed environmental and epidemiological data for more accurate outbreak forecasting and effective vector control strategies.