Comparative study of different machine learning models in landslide susceptibility assessment: A case study of Conghua District, Guangzhou, China

Abstract

Machine learning is currently one of the research hotspots in the field of landslide prediction. To clarify and evaluate the differences in characteristics and prediction effects of different machine learning models, Conghua District, which is the most prone to landslide disasters in Guangzhou, was selected for landslide susceptibility evaluation. The evaluation factors were selected by using correlation analysis and variance expansion factor method. Applying four machine learning methods namely Logistic Regression (LR), Random Forest (RF), Support Vector Machines (SVM), and Extreme Gradient Boosting (XGB), landslide models were constructed. Comparative analysis and evaluation of the model were conducted through statistical indices and receiver operating characteristic (ROC) curves. The results showed that LR, RF, SVM, and XGB models have good predictive performance for landslide susceptibility, with the area under curve (AUC) values of 0.752, 0.965, 0.996, and 0.998, respectively. XGB model had the highest predictive ability, followed by RF model, SVM model, and LR model. The frequency ratio (FR) accuracy of LR, RF, SVM, and XGB models was 0.775, 0.842, 0.759, and 0.822, respectively. RF and XGB models were superior to LR and SVM models, indicating that the integrated algorithm has better predictive ability than a single classification algorithm in regional landslide classification problems.

Keywords

Landslides susceptibility assessment / Machine learning / Logistic Regression / Random Forest / Support Vector Machines / XGBoost / Assessment model / Geological disaster investigation and prevention engineering

CRediT authorship contribution statement

Ao Zhang, Xing-yuezi Zhao, Xin-wen Zhao and Xiaozhan Zheng conceived of the presented idea. Ao Zhang, Xinwen Zhao and Xing-yuezi Zhao carried out the experiment. All authors discussed the results and contributed to the final manuscript.

Declaration of competing interest

The authors declare no conflicts of interest.

Acknowledgment

This research was supported by the projects of the China Geological Survey (DD20221729, DD20190291) and Zhuhai Urban Geological Survey (including informatization) (MZCD-2201-008). The authors are indebted to Guangzhou Municipal Bureau of Planning and Resources, Guangzhou Institute of Geological Survey, Guangzhou Urban Planning Survey and Design Institute for their assistance. The authors are also thankful to the reviewers and editors for their valuable comments and suggestions.

References

[1] Bennett ND, Croke BF, Guariso G, Guillaume JH, Hamilton SH, Jakeman AJ, Marsili-Libelli S, Newham LT, Norton JP, Perrin C, Pierce SA, Robson B, Seppelt R, Voinov AA, Fath BD, Andreassian V. 2013. Characterising performance of environmental models. Environmental modelling & software, 40, 1-20. doi: 10.1016/j.envsoft.2012.09.011.

[2] Cabrera AF. 1994. Logistic regression analysis in higher education:An applied perspective. In:Higher Education: Handbook of Theory and Research, 10, 225-256.

[3] Chen W, Chai HC, Zhao Z, Wang Q, Hong H. 2016. Landslide susceptibility mapping based on GIS andsupport vector machine models for the Qianyang County, China. Environmental Earth Sciences, 75(6), 1-13. doi: 10.1007/s12665-015-5093-0.

[4] Chen W, Chen X, Peng J B, Panahi M, Lee S. 2021. Landslide susceptibility modeling based on ANFIS with teaching-learningbased optimization and satin bowerbird optimizer. Geoscience Frontiers, 12(1), 93-107. doi: 10.1016/j.gsf.2020.07.012.

[5] Dou J, Yamagishi H, Pourghasemi HR, Yunus AP, Song X, Xu Y, Zhu Z. 2015. An integrated artificial neural network model for the landslide susceptibility assessment of Osado Island, Japan. Natural Hazards, 78(3), 1749-1776. doi: 10.1007/s11069-015-1799-2.

[6] Dou J, Yunus A P, Bui D T, Merghadi A, Sahana M,Zhu ZF, Chen, CW, Khosravi K, Yang Y, Pham BT. 2019. Assessment of advanced random forest and decision tree algorithms for modeling rainfallinduced landslide susceptibility in the Izu-Oshima Volcanic Is-land, Japan. Science of the Total Environment, 662, 332-346. doi: 10.1016/j.scitotenv.2019.01.221.

[7] Feng HJ, Zhou AG, Yu JY, Tang XM, Zheng JL, Chen XX, You SY. 2016. A comparative study on Plum-Rain-Triggered landslide susceptibility assessment models in West Zhejiang Province. Earth Science, 41(3), 403-415 doi: 10.3799/dgkx.2016.032. (in Chinese with English abstract).

[8] Guzzetti F, Mondini AC, Cardinali M, Fiorucci F, Santangelo M, Chang KT. 2012. Landslide inventory maps: New tools for an old problem. Earth-Science Reviews, 112(1-2), 42-66. doi: 10.1016/j.earscirev.2012.02.001.

[9] Hu T, Fan X, Wang S, Guo ZZ, Liu AC, Huang FM. 2020. Landslide susceptibility evaluation of Sinan County using logistics regression model and 3S technology. Bulletin of Geological Science and Technology, 39(2), 113-121 doi: 10.19509/j.cnki.dzkq.2020.0212. (in Chinese with English abstract).

[10] Huang FM, Chen JW, Du Z, Yao C, Huang JS, Jiang QH, Chang ZL, Li S. 2020. Landslide susceptibility pre-diction considering regional soil erosion based on machine-learning models. ISPRS International Journal of Geo-Information, 9(6), 377. doi: 10.3390/ijgi9060377.

[11] Huang FM, Hu SY, Yan XY, Li M, Wang JY, Li WB, Guo ZZ, Fan WY. 2022. Landslide susceptibility prediction and identification of its main environmental factors based on machine learning models. Bulletin of Geological Science and Technology, 41(2), 79-90 doi: 10.19509/j.cnki.dzkq.2021.0087. (in Chinese with English abstract).

[12] Huang Y, Zhao L. 2018. Review on landslide susceptibility mapping using support vector machines. Catena, 165, 520-529. doi: 10.1016/j.catena.2018.03.003.

[13] Jia YF, Wei WH, Chen W, Yang QZ, Sheng YF, Xu GL. 2023. Landslide susceptibility assessment based on the SOM-I-SVM model. Hydrogeology & Engineering Geology, 50(3), 125-137. doi: 10.16030/j.cnki.issn.1000-3665.202206041.

[14] Kanungo DP, Arora MK, Sarkar S, Gupta RP. 2006. A comparative study of conventional, ANN black box, fuzzy and combined neural and fuzzy weighting procedures for landslide susceptibility zonation in Darjeeling Himalayas. Engineering geology, 85(3-4), 347-366. doi: 10.1016/j.enggeo.2006.03.004.

[15] Lee S, Min K. 2001. Statistical analysis of landslide susceptibility at Yongin, Korea. Environmental geology, 40(9), 1095-1113. doi: 10.1007/s002540100310.

[16] Li JL, Ma DH, Wang W. 2016. Assessment of potential seismic landslide hazard based on evidence theory and entropy weight grey incidence. Journal of Central South University (Science and Technology), 47(5), 1730-1736 doi: 10.11817/j.issn.1672-7207.2016.05.036. (in Chinese with English abstract).

[17] Liu HH. 2012. The assessment of geohazard danger in Wenchuan County based on RS and GIS. Geology in China, 39(1), 243-251 (in Chinese with English abstract).

[18] Miao WD. 2003. Time prediction study on occurring of landslides in Bailuyuan, Xi'an. Northwestern Geology, 36(4), 90-95 (in Chinese with English abstract).

[19] Paraskevas T, Ioanna I. 2016. Landslide susceptibility mapping using a modified decision tree classifier in the Xanthi Perfection, Greece. Landslides, 13(2), 305-320. doi: 10.1007/s10346-015-0565-6.

[20] Peng CYJ, Lee K, Ingersoll GM. 2002. An introduction to logistic regression analysis and reporting. The Journal of Educational Research, 96(1), 3-14. doi: 10.1080/00220670209598786.

[21] Pham BT, Tien Bui D, Dholakia MB, Prakash I, Pham, HV. 2016. A comparative study of least square support vector machines and multiclass alternating decision trees for spatial prediction of rainfallinduced landslides in a tropical cyclones area. Geotechnical and Geological Engineering, 34, 1807-1824. doi: 10.1007/s10706-016-9990-0.

[22] Pham BT, Tien Bui D, Indra P, Dholakia M. 2015. Landslide susceptibility assessment at a part of Uttarakhand Himalaya, India using GIS-based statistical approach of frequency ratio method. nternational Journal of Engineering Research and Technology, 4(11), 338-344. doi: 10.17577/IJERTV4IS110285.

[23] Reichenbach P, Rossi M, Malamud BD, Mihir M, Guzzetti F. 2018. A review of statistically-based landslide susceptibility models. EarthScience Reviews, 180, 60-91. doi: 10.1016/j.earscirev.2018.03.001.

[24] Tien Bui D, Pradhan B, Lofman O, Revhaug I. 2012. Landslide susceptibility assessment in Vietnam using support vector machines, decision tree, and Naive Bayes models. Mathematical Problems in Engineering, 1-26. doi: 10.1155/2012/974638.

[25] Tien Bui D, Tuan TA, Klempe H, Pradhan B, Revhaug I. 2016. Spatial prediction models for shallow landslide hazards: A comparative assessment of the efficacy of support vector machines, artificial neural networks, kernel logistic regression, and logistic model tree. Landslides, 13, 361-378. doi: 10.1007/s10346-015-0557-6.

[26] Tsangaratos P, Ilia I, Hong H, Chen W, Xu C. 2017. Applying information theory and GIS-based quantitative methods to produce landslide susceptibility maps in Nancheng County, China. Landslides, 14(3), 1091-1111. doi: 10.1007/s10346-016-0769-4.

[27] Wang T, Liu JM, Li ZT, Xin P, Shi JS, Wu SR. 2021. Seismic landslide hazard assessment of China and its impact on national territory spatial planning. Geology in China, 48(1), 22-39 doi: 10.12029/gc20210102. (in Chinese with English abstract).

[28] Xia H, Yin KL, Liang X, Ma F. 2018. Landslide susceptibility assessment based on SVM-ANN Models: A case study for Wushan County in the Three Gorges Reservoir. The Chinese Journal of Geological Hazard and Control, 29(5), 13-19 doi: 10.16031/j.cnki.issn.1003-8035.2018.05.03. (in Chinese with English abstract).

[29] Xiong XH, Wang CL, Bai YJ, Tie YB, Gao YC, Li GH. 2022. Comparison of landslide susceptibility assessment based on multiple hybrid models at county level: A case study for Puge County, Sichuan Province. The Chinese Journal of Geological Hazard and Control, 33(4), 114-124. doi: 10.16031/j.cnki.issn.1003-8035.202202052.

[30] Yang DH, Fan W. 2015. Zoning of probable occurrence level of geological disasters based on ArcGIS--A case of Xunyang. The Chinese Journal of Geological Hazard and Control, 26(4), 82-86, 93 doi: 10.16031/j.cnki.issn.10038035.2015.04.14. (in Chinese with English abstract).

[31] Zêzere JL, Pereira S, Melo R, Oliveira SC, Garcia RAC. 2017. Mapping landslide susceptibility using data-driven methods. Science of the Total Environment, 589, 250-267. doi: 10.1016/j.scitotenv.2017.02.188.

[32] Zhang J, Yin KL, Wang JJ, Liu L, Huang FM. 2016. Study on landslide susceptibility evaluation for Wanzhou district of Three Gorges Reservoir. Chinese Journal of Rock Mechanics and Engineering, 35(2), 284-296 doi: 10.13722/j.cnki.jrme.2015.0318. (in Chinese with English abstract).