Abstract
The prediction of landslide is a complex task but preparing the landslide susceptibility map through artificial intelligence approaches can reduce life loss and damages resulting from landslides. The purpose of this study is to evaluate and compare the landslide susceptibility mapping (LSM) using six machine learning models, including random forest (RF), deep boost (DB), stochastic gradient boosting (SGB), rotation forest (RoF), boosted regression tree (BRT), and logit boost (LB) in the mountainous regions of western India. The landslide inventory map consisting of 184 landslide locations has been divided into two groups for training (70% data set) and validation (30% data set) purposes. Fourteen landslide triggering factors including slope, topographical roughness index, road density, topographical wetness index, elevation, slope length, drainage density, stream power index, geomorphology, rainfall, soil, lithology, lineament density, and normalized difference vegetation index have been considered using the boruta approach for the LSM. The results reveal that the RF model has the highest precision in terms of area under curve (0.88; 0.89), kappa (0.62; 0.50), accuracy (0.81; 0.77), and specificity (0.86; 0.86) both in the study region and secondary region, respectively. Hence, it can be concluded that the RF is an effective and promising technique as compared to DB, SGB, RoF, BRT, and LB for landslide susceptibility assessment in the research area as well as in regions having similar geo-environmental configuration.
Highlights
The frequency of natural disasters has been dramatically increased over the last few decades (UNISDR, 2015)
In congruence with the result of boruta algorithm, the result of multicollinearity analysis presented in Table 2, showed that the standard, plan and profile curvature were disqualified since there variance inflation factors (VIF)
It can be seen that the topographical factors such as slope, topographical wetness index (TRI), and elevation exerted more influence on landslide occurrences compared to the categorical factors in the study area
Summary
The frequency of natural disasters has been dramatically increased over the last few decades (UNISDR, 2015). A loss of billions of dollars and thousands of lives due to natural disasters, such as earthquakes, tsunamis, cyclones, floods, and landslides are recorded (Yesilnacar and Topal, 2005; Yilmaz, 2009). The occurrence of landslides constitutes about 17% of the natural disasters worldwide as reported by the Center for Research on the Epidemiology of Disasters (Pourghasemi et al, 2012; Nohani et al, 2019). Yilmaz (2009) and Pourghasemi et al (2012) believe that landslide events will tend to rise in the decades owing to deforestation, land-use change, and changing climatic condition. Many scientists have assessed landslide hazards and delineated the high-risk zones in past few decades
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
