基于CatBoost算法的滑坡敏感性模型——以毕节市大方县为例

Landslide Sensitivity Model Based on CatBoost Algorithm——Taking Dafang County of Bijie City as an Example

对山地地区建立滑坡敏感性模型,从而有效地避免地质灾害带来的风险,对促进地区经济发展和保证人民财产安全具有重要意义。以贵州省毕节市大方县为例,利用Borderline-SMOTE算法解决滑坡数据普遍存在的数据不平衡问题,进而运用CatBoost算法构建山体滑坡模型,最终选取查准率、召回率、F1-score和AUC作为模型评估指标进行对比。实验表明,利用经Borderline-SMOTE算法处理后的滑坡数据集构建的CatBoost模型相比较于传统机器学习模型的预测效果更好,尤其在AUC指标上效果提升显著,同时进一步说明了基于CatBoost算法的滑坡敏感性模型有着良好的分类效果。

Establishing a landslide sensitivity model for mountainous areas,so as to effectively avoid the risks caused by geological disasters,is of great significance to promote regional economic development and ensure the safety of people’s property.Taking Dafang County,Bijie City,Guizhou Province as an example,the Borderline-SMOTE algorithm is used to solve the data imbalance problem that is common in landslide data,and then the CatBoost algorithm is used to build a landslide model,and finally the precision rate,recall rate,F1-score and AUC are selected as model evaluation indicators are compared.The experimental results show that the CatBoost model constructed by using the landslide data set processed by the Borderline-SMOTE algorithm has better prediction effect than the traditional machine learning model,especially in the AUC index.The landslide sensitivity model has a good classification effect.