Automated Forest Fire Prediction Systems: A Comprehensive Review

Forest fires are one of the greatest risks to wildlife as well the human economy and the cases of wildfires have increased in the previous several years. The main causes are global warming and ongoing human intervention with natural environments. Automated forest fire prediction systems can lessen the harm caused to forests by detecting fire in the early stage and sending alerts to the concerned team so that appropriate preventive measures can be taken timely. In past few years, vast research has been conducted in this field while implementing various methodologies such as classification, clustering, supervised and unsupervised learning and so on. In this paper, the comprehensive review of such techniques is presented which can help users to identify the prediction systems based on its accuracy. This paper also presents a detailed comparative analysis of various existing techniques in terms of accuracy of the prediction systems.