A comprehensive evaluation model for forest fires based on MCDA and machine learning: A case study of Zhenjiang City, China

In recent years, forest fire accidents around the world have caused many casualties and property losses. Therefore, the development of a new risk model to assess the risk of forest fires is the main purpose at present. Zhenjiang City, China was set as the study area, and the original data around 2022 were applied to assess the forest fire risk. This study applied multi-criteria decision analysis technology and machine learning to construct a comprehensive assessment model. In addition, six methods were applied to evaluate the severity and display fire maps. Finally, four model parameters and receiver operating characteristic curves were applied to verify the accuracy of the model. The results show that the random forest has the best prediction effect. The accuracy and the area under the curve were 0.8086 and 0.9312, respectively. Moreover, land use, precipitation, and population density are the three most important factors. Therefore, the local government should strengthen the investigation of fire hazards and build a dynamic real-time natural disaster database.