Title: Optimizing the Predictive Ability of Machine Learning Methods for Landslide Susceptibility Mapping Using SMOTE for Lishui City in Zhejiang Province, China

Authors: Wang, Y (Wang, Yumiao); Wu, XL (Wu, Xueling); Chen, ZJ (Chen, Zhangjian); Ren, F (Ren, Fu); Feng, LW (Feng, Luwei); Du, QY (Du, Qingyun)

Source: INTERNATIONAL JOURNAL OF ENVIRONMENTAL RESEARCH AND PUBLIC HEALTH Volume: 16 DOI: 10.3390/ijerph16030368 published: FEB 2019  

Abstract: The main goal of this study was to use the synthetic minority oversampling technique (SMOTE) to expand the quantity of landslide samples for machine learning methods (i.e., support vector machine (SVM), logistic regression (LR), artificial neural network (ANN), and random forest (RF)) to produce high-quality landslide susceptibility maps for Lishui City in Zhejiang Province, China. Landslide-related factors were extracted from topographic maps, geological maps, and satellite images. Twelve factors were selected as independent variables using correlation coefficient analysis and the neighborhood rough set (NRS) method. In total, 288 soil landslides were mapped using field surveys, historical records, and satellite images. The landslides were randomly divided into two datasets: 70% of all landslides were selected as the original training dataset and 30% were used for validation. Then, SMOTE was employed to generate datasets with sizes ranging from two to thirty times that of the training dataset to establish and compare the four machine learning methods for landslide susceptibility mapping. In addition, we used slope units to subdivide the terrain to determine the landslide susceptibility. Finally, the landslide susceptibility maps were validated using statistical indexes and the area under the curve (AUC). The results indicated that the performances of the four machine learning methods showed different levels of improvement as the sample sizes increased. The RF model exhibited a more substantial improvement (AUC improved by 24.12%) than did the ANN (18.94%), SVM (17.77%), and LR (3.00%) models. Furthermore, the ANN model achieved the highest predictive ability (AUC = 0.98), followed by the RF (AUC = 0.96), SVM (AUC = 0.94), and LR (AUC = 0.79) models. This approach significantly improves the performance of machine learning techniques for landslide susceptibility mapping, thereby providing a better tool for reducing the impacts of landslide disasters.

Document Type: Article

Language: English

Authors’ key words: landslide susceptibility; Lishui City; machine learning; SMOTE; slope units; neighborhood rough set theory

Addresses of reprint authors:

[Du, QY] Wuhan Univ, Sch Resources & Environm Sci, Wuhan 430079, Hubei, Peoples R China

[Du, QY] Wuhan Univ, Key Lab GIS, Minist Educ, Wuhan 430079, Hubei, Peoples R China

[Du, QY] Wuhan Univ, Key Lab Digital Mapping & Land Informat Applicat, Natl Adm Surveying Mapping & Geoinformat, Wuhan 430079, Hubei, Peoples R China

[Du, QY] Wuhan Univ, Collaborat Innovat Ctr Geospatial Technol, Wuhan 430079, Hubei, Peoples R China

E-mail: wymfrank@whu.edu.cn; snowforesting@163.com; chen_cehui@163.com; renfu@whu.edu.cn; lwfeng@whu.edu.cn; qydu@whu.edu.cn

Addresses:

[Wang, Yumiao; Ren, Fu; Feng, Luwei; Du, Qingyun] Wuhan Univ, Sch Resources & Environm Sci, Wuhan 430079, Hubei, Peoples R China

[Wu, Xueling] China Univ Geosci, Inst Geophys & Geomat, Wuhan 430074, Hubei, Peoples R China

[Chen, Zhangjian] Zhejiang Acad Surveying & Mapping, Hangzhou 310012, Zhejiang, Peoples R China

[Ren, Fu; Du, Qingyun] Wuhan Univ, Key Lab GIS, Minist Educ, Wuhan 430079, Hubei, Peoples R China

[Ren, Fu; Du, Qingyun] Wuhan Univ, Key Lab Digital Mapping & Land Informat Applicat, Natl Adm Surveying Mapping & Geoinformat, Wuhan 430079, Hubei, Peoples R China

[Du, Qingyun] Wuhan Univ, Collaborat Innovat Ctr Geospatial Technol, Wuhan 430079, Hubei, Peoples R China

Impact factor：2.145