Evaluation of different machine learning models and novel deep learning-based algorithm for landslide susceptibility mapping

Abstract

The losses and damage caused by landslide are countless in the world every year. However, the existing approaches of landslide susceptibility mapping cannot fully meet the requirement of landslide prevention, and further excavation and innovation are also needed. Therefore, the main aim of this study is to develop a novel deep learning model namely landslide net (LSNet) to assess the landslide susceptibility in Hanyin County, China, meanwhile, support vector machine model (SVM) and kernel logistic regression model (KLR) were employed as reference model. The inventory map was generated based on 259 landslides, the training dataset and validation dataset were, respectively, prepared using 70% landslides and the remaining 30% landslides. The variance inflation factor (VIF) was applied to optimize each landslide predisposing factor. Three benchmark indices were used to evaluate the result of susceptibility mapping and area under receiver operating characteristics curve (AUROC) was used to compare the models. Result demonstrated that although the processing speed of LSNet model is the slowest, it still significantly outperformed its corresponding benchmark models with validation dataset, and has the highest accuracy (0.950), precision (0.951), F1 (0.951) and AUROC (0.941), which reflected excellent predictive ability in some degree. The achievements obtained in this study can improve the rapid response capability of landslide prevention for Hanyin County.

A review on landslide susceptibility mapping research in Bangladesh

“…This need can be met using machine and deep learning models for landslide susceptibility mapping. Machine and deep learning and also model hybridization have produced comparatively higher accuracy in landslide prediction across the world [ [61] , [62] , [63] ].…”

“…The use of machine learning algorithms in landslide susceptibility mapping is very limited in Bangladesh and no evidence was found for deep learning algorithms and physically-based methods [ 29 ]. Deep learning algorithm and their hybridization and ensembles are producing a very good results across the world [ [61] , [62] , [63] , [67] , [68] , [69] ]. In many tropical regions where rainfall is very dominant and shallow landslides are more common, different physically-based methods such as SHALSTAB, SINMAP, TRIGRS, SLIP etc.…”

Enhancing co-seismic landslide susceptibility, building exposure, and risk analysis through machine learning

“…Instability arises on slopes due to the self-weight of materials triggered by rainfall and strong ground motion 8 , 46 ; therefore, susceptibility is directly connected with the surrounding slopes; hence, the slope is one of the critical factors for assessing susceptibility 4 , 17 . The slope angle in the study area ranges from 0 to 87.98° in the study area considered.…”

An Adaptive Identification Method for Potential Landslide Hazards Based on Multisource Data

“…Landslides are one of the most common natural geological hazards, having significant socioeconomic impact worldwide and seriously threatening the safety of local people's lives and properties [1], as well as constraining local resource and economic development [2,3]. Due to the current development of geological surveys and evaluation, the identification of potential geological hazards, including landslides, is still not accurate or comprehensive.…”