Machine-Learning-Based Hybrid Modeling for Geological Hazard Susceptibility Assessment in Wudou District, Bailong River Basin, China

Abstract

In the mapping and assessment of mountain hazard susceptibility using machine learning models, the selection of model parameters plays a critical role in the accuracy of predicting models. In this study, we present a novel approach for developing a prediction model based on random forest (RF) by incorporating ensembles of hyperparameter optimization. The performance of the RF model is enhanced by employing a Bayesian optimization (Bayes) method and a genetic algorithm (GA) and verified in the Wudu section of the Bailong River basin, China, which is a typical hazard-prone, mountainous area. We identified fourteen influential factors based on field measurements to describe the “avalanche–landslide–debris flow” hazard chains in the study area. We constructed training (80%) and validation (20%) datasets for 378 hazard sites. The performance of the models was assessed using standard statistical metrics, including recall, confusion matrix, accuracy, F1, precision, and area under the operating characteristic curve (AUC), based on a multicollinearity analysis and Relief-F two-step evaluation. The results indicate that all three models, i.e., RF, GA-RF, and Bayes-RF, achieved good performance (AUC: 0.89~0.92). The Bayes-RF model outperformed the other two models (AUC = 0.92). Therefore, this model is highly accurate and robust for mountain hazard susceptibility assessment and is useful for the study area as well as other regions. Additionally, stakeholders can use the susceptibility map produced to guide mountain hazard prevention and control measures in the region.

Optimizing Random Forest for Debris Flow Susceptibility Assessment: A Comparative Study of Advanced Algorithms

“…Several researchers have initiated exploration into the application of optimization algorithms for the purpose of optimizing RF model (Jiang et al, 2022). Due to their robust search capabilities and features pertaining to global optimization, optimization algorithms such as genetic algorithm (GA) (Gu et al, 2023;Wang et al, 2023), whale optimization algorithm (WOA) (Zhao et al, 2022;Zhu et al, 2023), grey wolf optimizer (GWO) (Liu et al, 2021;Saha et al, 2022), and sparrow search algorithm (SSA) (Lin et al, 2022;Ma et al, 2023) have garnered significant attention. Additionally, the citation format has been corrected to adhere to academic standards.…”

“…The maximum depth of a decision tree is the highest number of levels of splits that the tree can have, extending from the root node to the leaf nodes. A greater maximum depth of the decision tree leads to enhanced model fitting capability, but it also elevates the potential for overfitting (Pradhan et al, 2021;Wang et al, 2023). The primary objective of this study is to identify the optimal combination of the number of decision trees and the maximum depth to enhance the predictability and utility of RF algorithm.…”

An Assessment of Geological Hazard Management using Dynamically Stabilized Recurrent Neural Network and Beluga Whale Optimization Algorithm

“…Iconic Buddhist cave temples such as Mogao Caves and Yungang Grottoes adorn its landscape, adding to its allure. [5,6]. Several geological hazards, including landslides, debris flows, collapses, and like are common in the area due to its unique environment, which includes manifold rainstorms, highest mountains, more faults.…”