Landslide Susceptibility Modeling: An Integrated Novel Method Based on Machine Learning Feature Transformation

Husam A H Al-Najjar,Biswajeet Pradhan,M Santosh,Bahareh Kalantar,Abdullah Alamri,Maher Ibrahim Sameen

doi:10.3390/rs13163281

Abstract

Landslide susceptibility modeling, an essential approach to mitigate natural disasters, has witnessed considerable improvement following advances in machine learning (ML) techniques. However, in most of the previous studies, the distribution of input data was assumed as being, and treated, as normal or Gaussian; this assumption is not always valid as ML is heavily dependent on the quality of the input data. Therefore, we examine the effectiveness of six feature transformations (minimax normalization (Std-X), logarithmic functions (Log-X), reciprocal function (Rec-X), power functions (Power-X), optimal features (Opt-X), and one-hot encoding (Ohe-X) over the 11conditioning factors (i.e., altitude, slope, aspect, curvature, distance to road, distance to lineament, distance to stream, terrain roughness index (TRI), normalized difference vegetation index (NDVI), land use, and vegetation density). We selected the frequent landslide-prone area in the Cameron Highlands in Malaysia as a case study to test this novel approach. These transformations were then assessed by three benchmark ML methods, namely extreme gradient boosting (XGB), logistic regression (LR), and artificial neural networks (ANN). The 10-fold cross-validation method was used for model evaluations. Our results suggest that using Ohe-X transformation over the ANN model considerably improved performance from 52.244 to 89.398 (37.154% improvement).

Highlights

We studied the influence of six feature transformation functions applied to three standard machine learning (ML) methods
Diverse approaches have been made in the last few years, demonstrating the success of several algorithms to enhance landslide assessment, including different architectures, factor optimizations, and training/sampling practices
Several feature transformation solutions were applied to three ML methods, such as XGB, logistic regression (LR), and artificial neural network (ANN)

Summary

Introduction

Landslides are among the common natural disasters that cause considerable damage to life and property in various regions around the globe, with governments and other agencies having to spend a substantial amount to monitor and mitigate the disaster [1,2]. Landslide-related damage cannot be entirely avoided, these events can be predicted to some extent so that damage to human life and property can be minimized [2]. Susceptibility maps are used to perform hazard and risk assessments in landslideprone areas. Landslide susceptibility assessment has been considered as a fundamental step in landslide risk management.

Methods

Results

Conclusion