Chunfang Kong^1,2,3,4, Kai Xu^1,2,3,*, Junzuo Wang¹, Yiping Tian^1,3,4, Zhiting Zhang^1,3,4 and Zhengping Weng^1,3,4, ¹School of Computer, China University of Geosciences, China, ²Hubei Key Laboratory of Intelligent Geo-Information Processing, China, ³Innovation Center of Mineral Resources Exploration Engineering Technology in Bedrock Area, Ministry of Natural Resources, China, ⁴National-Local Joint Engineering Laboratory on Digital Preservation and Innovative Technologies for the Culture of Traditional Villages and Towns, China

The random forest (RF) model is improved by the optimization of unbalanced geological hazards dataset, differentiation of continuous geological hazards evaluation factors, sample similarity calculation, and iterative method for finding optimal random characteristics by calculating out-of-bagger errors. The geological hazards susceptibility evaluation model based on optimized RF (OPRF) was established and used to assess the susceptibility for Lingyun County. Then, ROC curve and field investigation were performed to verify the efficiency for different geological hazards susceptibility assessment models. The AUC values for five models were estimated as 0.766, 0.814, 0.842, 0.846 and 0.934, respectively, which indicated that the prediction accuracy of the OPRF model can be as high as 93.4%. This result demonstrated that the geological hazards susceptibility assessment model based on OPRF has the highest prediction accuracy. Furthermore, the OPRF model could be extended to other regions with similar geological environment backgrounds for geological hazards susceptibility assessment and prediction.

Geological Hazards, Susceptibility Evaluation, Random Forest (RF), Optimized RF (OPRF), Geographical Information Systems (GIS).