摘要
以三峡库区秭归-巴东段为例,将地理加权回归(GWR)模型引入到研究区的空间尺度分割方法中,利用粒子群优化(PSO)算法对支持向量机(SVM)模型参数进行优化,构建GWR-PSO-SVM耦合模型,完成研究区滑坡易发性评价,并与传统的PSO-SVM耦合模型结果进行对比。结果表明,在特定类别精度分析、总体预测精度分析和曲线下面积分析中,本文方法评价效果均优于传统方法。
In this paper,we introduce the geographically weighted regression(GWR)into the spatial scale segmentation method of the study area,the particle swarm optimization(PSO)algorithm to optimize the parameters of support vector machine(SVM)model,and finally construct the GWR-PSO-SVM coupling model.In addition,the traditional PSO-SVM coupling model is constructed to compare with the new method.The results show that in the specific category accuracy analysis,the overall prediction accuracy analysis and the area under the curve analysis,the evaluation results of this method are better than the traditional methods.
作者
于宪煜
熊十力
YU Xianyu;XIONG Shili(School of Civil Engineering,Architecture and Environment,Hubei University of Technology,28 Nanli Road,Wuhan 430068,China)
出处
《大地测量与地球动力学》
CSCD
北大核心
2020年第2期187-192,共6页
Journal of Geodesy and Geodynamics
基金
国家自然科学基金(41807297)
湖北省教育厅科研计划中青年人才项目(Q20171410)~~
关键词
滑坡
滑坡易发性评价
空间尺度
空间尺度分割
地理加权回归
landslide
landslide susceptibility evaluation
spatial scale
spatial scale segmentation
geographically weighted regression
