Research on Numerical Simulation and Prevention Strategy of Geological Hazards by Integrating Machine Learning and GIS Technology

Abstract

- In order to prevent the occurrence of geologic disasters in a timely manner, and reduce the impact and threat of disasters on society. In this paper, firstly, GIS technology is used to collect data related to geological disasters and analyze the factors triggering geological disasters, so as to facilitate the subsequent establishment of geological disaster prediction models. Secondly, in order to make the collected data more accurate, an interactive iterative cleaning method is used to clean the data to ensure that the data will not affect the establishment of the model. Finally, machine learning is used to establish the geohazard prediction model to complete the prediction of geohazards. In the simulation test, the integration of machine learning and GIS technology disaster loss rate is lower, at about 15%. The predicted value is 0.867, and the discrete index score is around 0.426, which is more discrete. Therefore, combining machine learning and GIS technology can establish a more accurate prediction model of geologic disasters in order to reduce the harm of geologic disasters to human beings.