[1]艾 骁,张 健,付济民.不同特征选择方法于区域地震滑坡危险性预测结果的差异性分析——以汶川地震区为例[J].华南地震,2024,(02):39-51.[doi:10.13512/j.hndz.2024.02.06]
　AI Xiao,ZHANG Jian,FU Jimin.Difference in Regional Seismic Landslide Risk Prediction Results Based on Different Feature Selection Methods—A Case Study of Wenchuan Earthquake Area[J].,2024,(02):39-51.[doi:10.13512/j.hndz.2024.02.06]
点击复制

不同特征选择方法于区域地震滑坡危险性预测结果的差异性分析——以汶川地震区为例()
分享到：

华南地震[ISSN:1006-6977/CN:61-1281/TN]

卷:
期数:
2024年02期
页码:
39-51
栏目:
地震与地质灾害
出版日期:
2024-06-30

文章信息/Info

Title:
Difference in Regional Seismic Landslide Risk Prediction Results Based on Different Feature Selection Methods—A Case Study of Wenchuan Earthquake Area
文章编号:
1001-8662（2024）02-0039-13
作者:
艾 骁; 张 健; 付济民
黑龙江科技大学建筑工程学院，哈尔滨 150022
Author(s):
AI Xiao; ZHANG Jian; FU Jimin
School of Architecture and Civil Engineering，Heilongjiang University of Science and Technology， Harbin 150022， China
关键词:
汶川地震; 地震滑坡危险性; 主成分分析; Gini指数; 人工神经网络
Keywords:
Wenchuan earthquake;  Seismic landslide risk;  Principal component analysis;  Gini index;  Artificial neural network
分类号:
P694；
DOI:
10.13512/j.hndz.2024.02.06
文献标志码:
A
摘要:
区域性地震滑坡危险性评估模型是一个用于评定特定地区在地震发生时滑坡概率及其危害程度的关键工具。目前，以机器学习为代表的数学建模方法成为了构建该评估模型的主要手段。但是，由于影响因子自身的复杂性与多样性所产生的评估模型预测结果的差异性这一问题却少有研究。此次研究结合汶川地震区内11个影响因子，分别运用相关性系数、主成分分析及Gini指数三种特征选择方法形成三类数据集。结合人工神经网络模型构建了该区域地震滑坡危险性评估模型，并详细地分析了上述三类方法形成的数据集构建的评估模型于预测结果的差异性。结果表明：基于主成分分析法形成的数据集构建的评估模型，对于非常高危险性等级区域的划分精度最高，并且，频率比精度达到了92%，ROC曲线的预测精度达到了93.3%，预测精度均为三组评估模型中的最高值。此次研究旨在为相关研究人员在地震滑坡危险性评估模型的构建方面提供一定的思路，并为后续综合多个地震区、多组特征组成的不同维度的数据集构建一个具有普适性的特征选择方法提供一定的理论基础。
Abstract:
The regional seismic landslide risk assessment model is a key tool for evaluating the probability and sever?ity of landslides in specific areas when an earthquake occurs. Currently， machine learning-based mathematical modeling methods have become the primary means to construct the assessment model. However， limited research has been conducted on the difference in prediction results of the assessment model caused by the complex and di?verse nature of influencing factors. This study considered 11 influencing factors in the Wenchuan earthquake area and used three feature selection methods，namely correlation coefficient，principal component analysis，and Gini index，to create three types of datasets. Combined with the artificial neural network model，seismic landslide risk assessment models for the Wenchuan earthquake area were constructed based on different datasets obtained by the above three methods and the difference in the prediction results was meticulously analyzed. The results indicate that the assessment model based on the datasets obtained by the principal component analysis method achieves the high?est accuracy in identifying areas with a very high risk level. In addition，it demonstrates a frequency ratio accuracy of 92% and a prediction accuracy of the receiver operating characteristic (ROC) curve of 93.3%. Therefore，it ex?hibits the highest prediction accuracy among the three groups of assessment models. This research aims to provide valuable insights for researchers involved in the construction of seismic landslide risk assessment models. Addition?ally，it provides a theoretical basis for developing a universal feature selection method that integrates multidimen?sional datasets from multiple seismic regions and sets of features.

参考文献/References:

[1] Keefer，D K. Landslides caused by earthquake[J]. Geeologi?cal Society of America Bulletin，1984（95）：406-421.
[2] Guzzetti F，Reichenbach P，Ardizzone F，et al. Estimating the quality of landslide susceptibility models[J]. Geomorphology， 2006，81（1）：166-184.
[3] Hong H，Pradhan B，Jebur M N，et al. Spatial prediction of landslide hazard at the Luxi area（China）using support vec?tor machines[J]. Environmental Earth Sciences，2016，75 （1）：1-14.
[4] Ayalew L，Yamagishi H，Marui H，et al. Landslides in Sado Island of Japan：part II. GIS-based susceptibility mapping with comparisons of results from two methods and verifica?tions[J]. Engineering Geology，2005，81（4）：432-445.
[5] Begueira S. Validation and evaluation of predictive models in hazard assessment and risk management[J]. Natural Hazards， 2006，37（3）：315-329.
[6] Van Den Eeckhaut M，Herva's J. State of the art of national landslide databases in Europe and their potential for assess?ing landslide susceptibility，hazard and risk[J]. Geomorpholo?gy，2012（139）：545-558.
[7] Conforti M，Pascale S，Robustelli G，et al. Evaluation of pre?diction capability of the artificial neural networks for map?ping landslide susceptibility in the Turbolo River catchment （northern Calabria，Italy）[J]. Catena，2014（113）：236-250.
[8] Lee S，Ryu J H，Kim I S. Landslide susceptibility analysis and its verification using likelihood ratio，logistic regression， and artificial neural network models：case study of Youngin， Korea[J]. Landslides，2007，4（4）：327-338.
[9] Tien Bui D，Pradhan B，Lofman O，et al. Landslide suscepti?bility mapping at Hoa Binh province（Vietnam）using an adaptive neuro-fuzzy inference system and GIS[J]. Computers&Geosciences，2012，45（4）：199-211.
[10]侯敏，贾韶辉，郭兆成.基于RS与GIS的层次分析法在滑坡危险性评估中的应用——以四川宣汉天台乡滑坡为例[J].现代地质，2006（04）：668-672.
[11] Yalcin A. GIS-based landslide susceptibility mapping using analytical hierarchy process and bivariate statistics in Arde?sen（Turkey）：comparisons of results and confirmations[J]. Catena，2008，72（1）：1-12.
[12] Du G，Zhang Y，Zou L，et al. Co-seismic landslide hazard as?sessment of the 2017 MS 6.9 Milin earthquake，Tibet，China， combining the logistic regression-information value and Newmark displacement models[J/OL]. Bulletin of Engineer?ing Geology and the Environment，2022，81（10）：446[2023-09-10]. https://doi.org/10.1007/s10064-022-02901-x.
[13] Wang W，Liu Y，Huang Y. A machine learning approach for regional earthquake-induced landslide susceptibility assess?ment[J]. Landslides，2020，17（9），2147-2162.
[14] Jing L，Hudson J A. Numerical methods in rock mechanics [J]. International Journal of Rock Mechanics and Mining Sciences，2002，39（4）：409-427.
[15] Jing L. A review of techniques，advances and outstanding is?sues in numerical modelling for rock mechanics and rock engineering[J]. International Journal of Rock Mechanics and Mining Sciences，2003，40（3）：283-353.
[16] Kanungo D P，Arora M K，Sarkar S，et al. A comparative study of conventional，ANN black box，fuzzy and combined neural and fuzzy weighting procedures for landslide suscep?tibility zonation in Darjeeling Himalayas[J]. Engineering Geology，2006，85（3-4）：347-366.
[17] Choi J，Oh H J，Lee H J，et al. Combining landslide suscep?tibility maps obtained from frequency ratio，logistic regres?sion，and artificial neural network models using ASTER im?ages and GIS[J]. Engineering Geology，2012（124）：12-23.
[18] Xu C，Xu X W，Dai F C，et al. Comparison of different mod?els for susceptibility mapping of earthquake triggered land?slides related with the 2008 Wenchuan earthquake in China [J]. Computers&Geosciences，2012（46）：317-329.
[19] Aditian A，Kubota T，Shinohara Y. Comparison of GIS-based landslide susceptibilitymodels using frequency ratio， logistic regression，and artificial neural network in a tertia?ry region of Ambon，Indonesia[J]. Geomophology. 2018 （318）：101-111.
[20]许冲.汶川地震滑坡详细编录及其与全球其他地震滑坡事件对比[J].科技导报，2012，30（25）：18-26．
[21]许冲，徐锡伟，吴熙彦，等. 2008年汶川地震滑坡详细编目及其空间分布规律分析[J].工程地质学报，2013，21 （1）：1-20．
[22] Guzzetti F，Carrarra A，Cardinali M，et al. Landslide hazard evaluation：a review of current techniques and their appli?cation in a multi-scale study[J]. Central Italy. Geomorpholo?gy，1999（31）：181-216.
[23] Lineback Gritzner M，Marcus W A，Aspinall R，et al. As?sessing landslide potential using GIS，soil wetness modeling and topographic attributes，Payette River，Idaho[J]. Geomor?phology，2001，37（1/2）：149-165.
[24] Ayalew L，Yamagishi H. The application of GIS-based logis?tic regression for landslide susceptibility mapping in the Kakuda-Yahiko Mountains，Central Japan[J]. Geomorpholo?gy，2005，65（1/2）：15-31.
[25] Xu C，Xu X. Controlling parameter analyses and hazard mapping for earthquake-triggered landslides：an example from a square region in Beichuan County，Sichuan Prov?ince，China[J]. Arabian Journal of Geosciences，2013，6 （10）：3827-3839.
[26] Dai F，Lee C F. Landslide characteristics and slope instabili?ty modeling using GIS，Lantau Island，Hong Kong[J]. Geo?morphology，2002，42（3/4）：213-228.
[27]胡聿贤.地震工程学（第二版）[M].北京：地震出版社， 2009.
[28]许冲，戴福初，姚鑫，等.基于GIS与确定性系数分析方法的汶川地震滑坡易发性评价[J].工程地质学报，2010，18 （1）：15-26．
[29] Regmi A D，Devkota K C，Yoshida K，et al. Application of frequency ratio，statistical index，and weights of evidence models and their comparison in landslide susceptibility mapping in Central Nepal Himalaya[J]. Arabian Journal of Geosciences，2014，7（2）：725-742.
[30] Pachauri A K，Pant M. Landslide hazard mapping based on geological attributes[J]. Engineering Geology，1992（ 32）：81-100.
[31] Luzi L，Pergalani F. Slope instability in static and dynamic conditions for urban planning：the"Oltre Po Pavese"case history（Region Lombardia-Italy）[J]. Natural Hazards， 1999，20（1）：57-82.
[32] Uromeihy A，Mahdavifar M R. Landslide hazard zonation of the Khorshrostam area，Iran[J]. Bulletin of Engineering Ge?ology and the Environment，2000，58（3）：207-213.
[33] Saha A K，Gupta R P，Arora M K. GIS-based landslide haz?ard zonation in the Bhagirathi（Ganga）valley，Himalayas[J]. International Journal of Remote Sensing，2002，23（2）：357-369.
[34]李芸芸.地震滑坡危险性初步研究及城市地震灾害三维场景模拟新方法[D].北京：中国地震局工程力学研究所， 2016.
[35]艾骁.基于机器学习的地震滑坡易发性评估模型构建[D].北京：中国地震局工程力学研究所，2021.
[36] McCulloch W S，Pitts W. A logical calculus of the ideas im?manent in nervous activity[J]. The Bulletin of Mathematical Biophysics，1990，52（1/2）：99-115.
[37] Rosenblatt F. The perceptron：aprobabilistic model for infor?mation storage and organization in the brain[J]. Psychologi?cal Review，1958（65）：386-408.
[38] Yesilnacar E，Topal T. Landslide susceptibility mapping：a comparison of logistic regression and neural networks meth?ods in a medium scale study，Hendek region（Turkey）[J]. Engineering Geology，2005，79（3/4）：251-266.
[39] Huang F，Yao C，Liu W，et al. Landslide susceptibility as?sessment in the Nantian area of China：a comparison of fre?quency ratio model and support vector machine[J]. Geomat?ics，Natural Hazards and Risk，2018，9（1）：919-938.
[40] Chang Z，Du Z，Zhang F，et al. Landslide susceptibility pre?diction based on remote sensing images and gis：compari?sons of supervised and unsupervised machine learning mod?els[J]. Remote Sensing，2020，12（3）：502.
[41] Thiery Y，Malet J P，Sterlacchini S，et al. Landslide suscepti?bility assessment by bivariate methods at large scales：appli?cation to a complex mountainous environment[J]. Geomor?phology，2007，92（1）：38?59.
[42] Tom F. An introduction to ROC analysis[J]. Pattern Recogni?tion Letters，2006，27（8）：861?874.
?

备注/Memo

备注/Memo:
收稿日期：2023-10-10
基金项目：黑龙江科技大学引进高层次人才科研启动基金项目（HKD202133）；2022年度黑龙江省省属高等学校基本科研业务费科研项目（2022-KYYWF-0560）联合资助。
作者简介：艾骁（1988-），男，博士研究生，讲师，研究方向为地学空间建模、地震地质灾害预警技术。E-mail：axz.1988@163.com
更新日期/Last Update: 2024-06-30