PSO Trained Neural Networks for Predicting Forest Fire Size: A Comparison of Implementation and Performance

Forest fires are a dangerous and devastating phenomenon. Being able to accurately predict the burned area of a forest fire could potentially limit human casualties as well as better prepare for the ensuing economical and ecological damage. A data set from the Montesinho Natural Park in Portugal provides a difficult regression task regarding the prediction of forest fire burn area due to the limited amount of data entries and the right skew nature of the data set. This paper shows how the use of a novel input structure and representation of the data, along with using Particle Swarm Optimization (PSO) instead of Backpropagation to determine weights of an Artificial Neural Network (ANN), improves error rates significantly.