Investigation of landslide susceptibility using machine learning techniques in a mountainous region.

Thesis Abstract

Abstract
Landslides are a significant natural hazard that poses risks to human lives, infrastructure, and the environment, particularly in mountainous regions. Traditional methods of assessing landslide susceptibility often rely on manual interpretation of geological and topographical data, which can be time-consuming and subjective. In recent years, machine learning techniques have emerged as powerful tools for analyzing complex datasets and predicting landslide susceptibility with higher accuracy. This thesis presents an investigation into the application of machine learning techniques for assessing landslide susceptibility in a mountainous region. The study begins with a comprehensive review of existing literature on landslides, machine learning, and their intersection. The literature review highlights the limitations of traditional methods and the potential of machine learning algorithms to improve landslide susceptibility mapping. Various machine learning algorithms, including decision trees, support vector machines, and neural networks, are discussed in relation to their suitability for landslide susceptibility analysis. The research methodology section outlines the steps taken to collect and preprocess geospatial data, including topographic, geological, and land cover information. Feature selection techniques are employed to identify the most relevant variables for predicting landslide susceptibility. The machine learning models are trained and tested using historical landslide data to evaluate their performance and identify the most accurate algorithm for the study area. The findings of the study reveal the effectiveness of machine learning techniques in predicting landslide susceptibility in the mountainous region. The selected machine learning model demonstrates high accuracy in identifying areas at high risk of landslides, providing valuable insights for land use planning and disaster mitigation efforts. The discussion delves into the factors influencing landslide susceptibility, such as slope gradient, soil type, and land cover, and how these variables are captured by the machine learning model. In conclusion, this thesis contributes to the field of geoscience by showcasing the potential of machine learning techniques in assessing landslide susceptibility. The study highlights the importance of leveraging advanced technologies to enhance landslide risk assessment and improve disaster preparedness in mountainous regions. Recommendations for future research include exploring the integration of remote sensing data and real-time monitoring systems to further enhance landslide prediction accuracy. Keywords Landslide susceptibility, Machine learning, Geospatial analysis, Mountainous region, Disaster risk assessment. (Word count 291)