Time Series Modelling of Landslides using Reinforcement Learning for Feature Selection
Abstract
Landslides pose significant risks to human life, the community, and the environment, yet their prediction remains a complex and unexplored challenge. Existing prediction models often rely on surface measurements and satellite images, neglecting the critical role of providing deeper insights into landslide analysis. The literature review highlights a lack of research in time series decomposition techniques, despite their potential to improve prediction accuracy. Similarly, feature selection methods that enhance model robustness and precision have not been adequately addressed. This study presents a novel approach to predicting landslide displacement by combining feature selection through reinforcement learning techniques with advanced time-series machine learning modeling techniques. Reinforcement learning is used to dynamically select impactful features, optimizing the input for prediction models. These insights guide the development of advanced and hybrid machine learning models trained and tested on comprehensive datasets, aiming to enhance prediction accuracy and efficiency. Despite its complexity, the hybrid Long Short-Term Memory (LSTM) and Convolutional Neural Networks (CNN) model achieved a lower accuracy (61.70%) compared to traditional machine learning models such as Logistic Regression (89.06%), Support Vector Classifier (88.50%), Decision Tree Classifier (96.68%), Extra Tree Classifier (96.67%), Random Forest Classifier (97.31%), and XGBoost (96.25%). This work sets a new benchmark for predictive modeling and operational strategies in landslide monitoring by addressing key field gaps.
