2025 Vol. 52, No. 1
Article Contents

FENG Minxuan, MAO Yimin, JIA Jun, QI Qi, MENG Xiaojie, LIU Gang, GAO Bo, GAO Manxin. 2025. Comparison of the advantages and disadvantages of different machine learning methods in geohazard risk assessment: Taking Da'an Town, Ningqiang County as an example[J]. Geology in China, 52(1): 205-214. doi: 10.12029/gc20231018003
Citation: FENG Minxuan, MAO Yimin, JIA Jun, QI Qi, MENG Xiaojie, LIU Gang, GAO Bo, GAO Manxin. 2025. Comparison of the advantages and disadvantages of different machine learning methods in geohazard risk assessment: Taking Da'an Town, Ningqiang County as an example[J]. Geology in China, 52(1): 205-214. doi: 10.12029/gc20231018003

Comparison of the advantages and disadvantages of different machine learning methods in geohazard risk assessment: Taking Da'an Town, Ningqiang County as an example

    Fund Project: Supported by the projects of China Geological Survey (No.20230436, No.DD20221739) and Natural Resources Shaanxi Satellite Application Technology Center (No.SCZK2022–CS–1645/001).
More Information
  • Author Bio: FENG Minxuan, female, born in 1990, assistant researcher, mainly engaged in geological survey and InSAR technology application; E-mail: fengminxuan@mail.cgs.gov.cn
  • Corresponding author: QI Qi, male, born in 1989, engineer, mainly engaged in the research work of engineering geology, structural geology and other aspects; E-mail: xqq8901@163.com
  • This paper is the result of geohazard survey engineering.

    Objective

    The occurrence of geohazards are influenced by various factors, which have uncertainty and complexity, making it difficult to assess the risk of geohazards. With the development of AI technology, intelligent algorithms can more accurately calculate the complex and nonlinear relationships between geohazard triggering indexes, greatly improving the accuracy of geological hazard risk assessment models.

    Methods

    Based on the field geological survey data of Da'an Town, Ningqiang County, 12 indexes closely related to the occurrence of geohazards were selected, namely elevation, slope, slope height, slope direction, slope type, engineering geological rock formations, fault distance, water system distance, road distance, vegetation coverage, rainfall, and seismic ground motion, as risk zoning evaluation factors. By constructing a sample set, Bayesian, strategy gradient neural network, random forest, KNN and neural network algorithm are used to model and compare the geohazard risk assessment result in Da'an Town, Ningqiang County.

    Results

    The experimental results show that the Bayesian model (AUC 0.894) performs the best, with the vast majority of geohazards located in the extremely high and high−risk evaluated areas, and meets the requirements for prediction accuracy evaluation.

    Conclusions

    It is feasible to choose Bayesian algorithm models for geological hazard risk assessment when the number of geohazard samples is small.

  • 加载中
  • [1] Cao W G, Fu Y, Dong Q Y, Wang H G, Ren Y, Li Z Y, Du Y Y. 2023. Landslide susceptibility assessment in Western Henan Province based on a comparison of conventional and ensemble machine learning[J]. China Geology, 6(3): 409−419.

    Google Scholar

    [2] Cao Puyuan, Qiu Haijun, Hu Sheng, Yang Dongdong. 2017. Research on scale parameter frequency distribution of regional collapse and landslide in Ningqiang County[J]. Journal of Catastrophology, 32(4): 126−131 (in Chinese with English abstract).

    Google Scholar

    [3] Chen Shuiman, Zhao Huilong, Xu Zhen, Xie Wei, Liu Liang, Li Quanyue. 2022. Landslide risk assessment in Nanping City based on artificial neural networks model[J]. The Chinese Journal of Geological Hazard and Control, 33(2): 133−140 (in Chinese with English abstract).

    Google Scholar

    [4] Dou Jie, Xiang Zilin, Xu Qiang, Zheng Penglin, Wang Xiekang. Su Aijun, Liu Junqi, Luo Wanqi. 2023. Application and development trend of machine learning in landslide intelligent disaster prevention and mitigation[J]. Earth Science, 48(5): 1657−1674 (in Chinese with English abstract).

    Google Scholar

    [5] Fawcett T. 2006. An introduction to ROC analysis[J]. Pattern Recognition Letters, 27(8): 861−874. doi: 10.1016/j.patrec.2005.10.010

    CrossRef Google Scholar

    [6] Fang Ranke, Liu Yanhui, Huang Zhiquan. 2021. A review of the methods of regional landslide hazard assessment based on machine learning[J]. The Chinese Journal of Geological Hazard and Control, 32(4): 1−8 (in Chinese with English abstract).

    Google Scholar

    [7] Li Guanghui, Tie Yongbo. 2023. Comparative study on modeling methods of comprehensive geological hazard susceptibility based on information model[J]. Journal of Catastrophology, 38(3): 212−221 (in Chinese with English abstract).

    Google Scholar

    [8] Li Jiahao, Xie Wanli, Yan Ming, Liu Qiqi, He Gaorui. 2023. Research on geological hazard risk assessment based on PCA and improved AHP–CRITIC method: A case study of Shenmu, Shaanxi Province[J]. Journal of Earth Environment, 14(4): 472−487 (in Chinese with English abstract).

    Google Scholar

    [9] Li Ming, Jiang Weijun, Dong Jiahui, Jin Shaofeng, Zhang Chenwei, Niu Ruiqing. 2023. Evaluation of landslide hazards susceptibility based on machine learning: Taking the Three Gorges reservoir area as an example[J]. South China Geology, 39(3): 413−427 (in Chinese with English abstract).

    Google Scholar

    [10] Li Xin, Xue Guicheng, Liu Changzhu, Xia Nan, Yang Yongpeng, Yang Feng, Wang Xiaolin, Chang Zhenyu. 2022. Evaluation of geohazard susceptibility based on information value model and information value–logistic regression model: A case study of the central mountainous area of Hainan Island[J]. Journal of Geomechanics, 28(2): 294−305 (in Chinese with English abstract).

    Google Scholar

    [11] Ma Xiao, Wang Nianqin, Li Xiaokang, Yan Dong, Li Jialin. 2022. Assessment of landslide susceptibility based on RF–FR model: Taking Lueyang County as an example[J]. Northwestern Geology, 55(3): 335−344 (in Chinese with English abstract).

    Google Scholar

    [12] Meng Xiaojie, Zhang Xinshe, Zeng Qingming, Wang Dong. 2022. The susceptibility evaluation of loess landslide based on weighted information value method—Taking 1: 50000 map of Maiji District of Tianshui City as an example[J]. Northwestern Geology, 55(2): 249−259 (in Chinese with English abstract).

    Google Scholar

    [13] Mao Y M, Mwakapesa D S, Wang G L, Nanehkaran Y A, Zhang M S. 2021. Landslide susceptibility modelling based on AHC–OLID clustering algorithm[J]. Advances in Space Research, 68(1): 301−316. doi: 10.1016/j.asr.2021.03.014

    CrossRef Google Scholar

    [14] Pourghasemi H R, Rahmati O. 2018. Prediction of the landslide susceptibility: Which algorithm, which precision?[J]. Catena, 162: 177−192. doi: 10.1016/j.catena.2017.11.022

    CrossRef Google Scholar

    [15] Qiu Haijun, Cao Mingming, Liu Wen, Hao Junqing, Wang Yanlin. 2014. Research on the spatial point pattern of geohazard: A case of Ningqiang County[J]. Journal of Arid Land Resources and Environment, 28(3): 107−111 (in Chinese with English abstract).

    Google Scholar

    [16] Shirzadi A, Solaimani K, Roshan M H, Kavian A, Chapi K, Shahabi H, Keesstra S, Ahmad B B, Bui D T. 2019. Uncertainties of prediction accuracy in shallow landslide modeling: Sample size and raster resolution[J]. Catena, 178: 172−188. doi: 10.1016/j.catena.2019.03.017

    CrossRef Google Scholar

    [17] Sun Jianfeng, Ma Chao, Hu Jinshu, Yan Tiesheng, Gao Jiajun, Xu Hui. 2023. Susceptibility evaluation of geological hazard by coupling grey relational degree and analytic hierarchy process: A case of Chongtou Town, Yunhe County, Zhejiang Province[J]. Journal of Engineering Geology, 31(2): 538−551 (in Chinese with English abstract).

    Google Scholar

    [18] Tang Yaming, Zhang Maosheng. 2011. Landslide risk assessment difficulties and methods: A review[J]. Hydrogeology and Engineering Geology, 38(2): 130−134 (in Chinese with English abstract).

    Google Scholar

    [19] Tamura R, Kobayashi K, Takano Y, Miyashiro R, Nakata K, Matsui T. 2019. Mixed integer quadratic optimization formulations for eliminating multicollinearity based on variance inflation factor[J]. Journal of Global Optimization, 73: 431−446. doi: 10.1007/s10898-018-0713-3

    CrossRef Google Scholar

    [20] Wang Bendong, Li Siquan, Xu Wanzhong, Yang Yong, Li Yongyun. 2024. A comparative study of landslide susceptibility evaluation based on three different machine learning algorithms[J]. Northwestern Geology, 57(1): 34−43 (in Chinese with English abstract).

    Google Scholar

    [21] Wu Shuren, Shi Jusong, Wang Tao, Zhang Chunshan, Shi Ling. 2012. The Theory and Technology of Landslide Risk Assessment[M]. Beijing: Science Press (in Chinese).

    Google Scholar

    [22] Xue Qiang, Zhang Maosheng, Dong Ying, Meng Xiaojie, Guo Xiaopeng, Feng Wei, Hong Bo, Wang Tao, Liu Wenhui, Tian Zhongying, Zhang Ge, Lu Na. 2023. Refinement risk identification of loess geo–hazards based on DEM and remote sensing—Taking Mizhi County in the Loess Plateau of Northern Shaanxi as an example[J]. Geology in China, 50(3): 926−942 (in Chinese with English abstract).

    Google Scholar

    [23] Yao Xiaoyue, Su Wenji, Li Xiujuan, Zheng Zhiwen, Mei Weibiao. 2023. Risk assessment of geological disasters in low mountain and hilly regions based on multiple combined models and its accuracy analysis[J]. South China Journal of Seismology, 43(3): 95−109 (in Chinese with English abstract).

    Google Scholar

    [24] Zhang Linfan, Wang Jiayun, Zhang Maosheng, Chen Shebin, Wang Tao. 2022. Evaluation of regional landslide susceptibility assessment based on BP neural network[J]. Northwestern Geology, 55(2): 260−270 (in Chinese with English abstract).

    Google Scholar

    [25] Zhang Maosheng, Xue Qiang, Jia Jun, Xu Jiwei, Gao Bo, Wang Jiayun. 2019. Methods and practices for the investigation and risk assessment of geo–hazards in mountainous towns[J]. Northwestern Geology, 52(2): 125−135 (in Chinese with English abstract).

    Google Scholar

    [26] Zhang Wenlong, Zhang Zhenkai, Yang Shuai. 2023. Study on intelligent evaluation and zoning of geohazards risk in Mianluening area[J]. Northwestern Geology, 56(1): 276−283 (in Chinese with English abstract).

    Google Scholar

    [27] Zhang A, Zhao X W, Zhao X Y, Zheng X Z, Zeng M, Huang X, Wu P, Jiang T, Wang S C, He J, Li Y Y. 2024. Comparative study of different machine learning models in landslide susceptibility assessment: A case study of Conghua District, Guangzhou, China[J]. China Geology, 7(1): 104−115.

    Google Scholar

    [28] Zhou Jingjing, Zhang Xiaomin, Zhao Fasuo, Li Hui, Liu Hainan. 2019. Research on risk assessment of geological hazards in Qinling–Daba mountain area, south Shaanxi Province[J]. Journal of Geomechanics, 25(4): 544−553 (in Chinese with English abstract).

    Google Scholar

    [29] 曹璞源, 邱海军, 胡胜, 杨冬冬. 2017. 区域崩塌和滑坡规模参数频率分布研究—以秦巴山地宁强县为例[J]. 灾害学, 32(4): 126−131.

    Google Scholar

    [30] 陈水满, 赵辉龙, 许震, 谢伟, 刘亮, 李全悦. 2022. 基于人工神经网络模型的福建南平市滑坡危险性评价[J]. 中国地质灾害与防治学报, 33(2): 133−140.

    Google Scholar

    [31] 窦杰, 向子林, 许强, 郑鹏麟, 王协康, 苏爱军, 刘军旗, 罗万祺. 2023. 机器学习在滑坡智能防灾减灾中的应用与发展趋势[J]. 地球科学, 48(5): 1657−1674.

    Google Scholar

    [32] 方然可, 刘艳辉, 黄志全. 2021. 基于机器学习的区域滑坡危险性评价方法综述[J]. 中国地质灾害与防治学报, 32(4): 1−8.

    Google Scholar

    [33] 李光辉, 铁永波. 2023. 基于信息量模型的综合地质灾害易发性[J]. 灾害学, 38(3): 212−221.

    Google Scholar

    [34] 李嘉昊, 谢婉丽, 严明, 刘琦琦, 何高锐. 2023. 基于PCA与改进AHP–CRITIC法的地质灾害风险评价研究—以神木市为例[J]. 地球环境学报, 14(4): 472−487.

    Google Scholar

    [35] 李明, 蒋委君, 董佳慧, 金少锋, 张宸伟, 牛瑞卿. 2023. 基于机器学习的滑坡灾害易发性评价—以三峡库区为例[J]. 华南地质, 39(3): 413−427.

    Google Scholar

    [36] 李信, 薛桂澄, 柳长柱, 夏南, 杨永鹏, 杨峰, 王晓林, 常振宇. 2022. 基于信息量模型和信息量–逻辑回归模型的海南岛中部山区地质灾害易发性研究[J]. 地质力学学报: 28(2): 294–305.

    Google Scholar

    [37] 马啸, 王念秦, 李晓抗, 严冬, 李嘉琳. 2022. 基于RF–FR模型的滑坡易发性评价—以略阳县为例[J]. 西北地质, 55(3): 335−344.

    Google Scholar

    [38] 孟晓捷, 张新社, 曾庆铭, 王冬. 2022. 基于加权信息量法的黄土滑坡易发性评价—以1∶5万天水市麦积幅为例[J]. 西北地质, 55(2): 249−259.

    Google Scholar

    [39] 邱海军, 曹明明, 刘闻, 郝俊卿, 王雁林. 2014. 区域地质灾害的空间点格局分析研究—以宁强县为例[J]. 干旱区资源与环境, 28(3): 107−111. doi: 10.3969/j.issn.1003-7578.2014.03.019

    CrossRef Google Scholar

    [40] 孙剑锋, 马超, 胡金树, 闫铁生, 杲加俊, 徐辉. 2023. 基于灰色关联度与层次分析法耦合的地质灾害易发性评价—以浙江省云和县崇头镇为例[J]. 工程地质学报, 31(2): 538−551.

    Google Scholar

    [41] 唐亚明, 张茂省. 2011. 滑坡风险评价难点及方法综述[J]. 水文地质工程地质, 38(2): 130−134.

    Google Scholar

    [42] 王本栋, 李四全, 许万忠, 杨勇, 李永云. 2024. 基于3种不同机器学习算法的滑坡易发性评价对比研究[J]. 西北地质, 57(1): 34−43.

    Google Scholar

    [43] 吴树仁, 石菊松, 王涛, 张春山, 石玲. 2012. 滑坡风险评估理论与技术[M]. 北京: 科学出版社.

    Google Scholar

    [44] 薛强, 张茂省, 董英, 孟晓捷, 郭小鹏, 冯卫, 洪勃, 王涛, 刘文辉, 田中英, 张戈, 卢娜. 2023. 基于DEM和遥感的黄土地质灾害精细化风险识别—以陕北黄土高原区米脂县为例[J]. 中国地质, 50(3): 926−942. doi: 10.12029/gc20220801001

    CrossRef Google Scholar

    [45] 姚小月, 宿文姬, 李秀娟, 郑志文, 梅伟标. 2023. 基于多种组合模型的低山丘陵区地质灾害危险性评价及精度分析[J]. 华南地震, 43(3): 95−109.

    Google Scholar

    [46] 张林梵, 王佳运, 张茂省, 陈社斌, 王涛. 2022. 基于BP神经网络的区域滑坡易发性评价[J]. 西北地质, 55(2): 260−270.

    Google Scholar

    [47] 张茂省, 薛强, 贾俊, 徐继维, 高波, 王佳运. 2019. 山区城镇地质灾害调查与风险评价方法及实践[J]. 西北地质, 52(2): 125−135.

    Google Scholar

    [48] 张文龙, 张振凯, 杨帅. 2023. 勉略宁地区地质灾害危险性智能评价和区划研究[J]. 西北地质, 56(1): 276−282.

    Google Scholar

    [49] 周静静, 张晓敏, 赵法锁, 李辉. 2019. 陕南秦巴山区地质灾害危险性评价研究[J]. 地质力学学报, 25(4): 544−553. doi: 10.12090/j.issn.1006-6616.2019.25.04.053

    CrossRef Google Scholar

  • 加载中
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(5)

Tables(1)

Article Metrics

Article views(99) PDF downloads(13) Cited by(0)

Access History

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint