ARTIFICIAL INTELLIGENCE OPTIMIZATION FOR FOREST FIRE RISK PREDICTING APPLIED TO GREEN ENVIRONMENT

Abstract

This paper aims to contribute to the field of green environment and meteorological signal processing by exploring methods for analyzing and predicting environmental factors. The focus is on developing an intelligent approach for preventing and predicting wildfires that may arise due to changes in atmospheric temperature or other conditions. The proposed solution involves using machine learning and evolutionary deep learning to create a neural model that can interact with the Internet of Things (IoT) and respond in real-time to minimize potential damage. The experiments were carried out in the forests of Jandouba, Tunisia, using Python software. The results demonstrate that this approach offers significant advantages over the most ranked existing Canadian method (FWI) for fire weather index.