Evaluating model performance for spatial assessments of wildfire ignition risk. Contribution to Deliverable 2.5a for the Project D5-2 Climate Change Impacts on Natural Capital. The James Hutton Institute, Aberdeen, Scotland.
This work aims to evaluate the performance of a machine learning model for assessing wildfire ignition risk using independent validation.
Used an independent validation approach to assess model performance.
Conducted spatial assessments of wildfire ignition risk using machine learning techniques.
The model effectively identifies regions at higher risk for wildfire ignition.
Findings suggest improved accuracy and reliability in wildfire risk assessments.
Abstract
The purpose of this work was to use an independent validation approach to evaluate the performance of the Machine Learning (ML) model for conducting spatial assessments of wildfire ignition risk.