Researchers use machine learning to predict large wildfires

A multidisciplinary group of researchers from the University of California, Irvine created a machine learning model to predict the potential of large wildfires from the time of ignition. The decision classifier model uses a single dataset to predict whether a fire will be large roughly 50% of the time, outperforming more complex models tested by researchers that rely on multiple weather variables. Researchers trained the AI with air moisture data over the span of the average weather report using data from more than 1,100 fire events in Alaska between 2001 and 2017 from the Alaska Large Fire Database. Each fire was then labeled as small, medium, or large. The model is then able to predict approximately 40% of ignitions that lead to large wildfires that account for 75% of burned area during that time period.