Supporting data for "Machine learning for natural hazard data analyses and data-driven geotechnical engineering applications"

Extensive exploration of novel machine learning (ML) technologies within the domain of geoscience has been actively pursued. However, the integration of ML with geoscience is in its early stages and unevenly advancing. My Ph.D. thesis entitled "Machine learning for natural hazard data analyses and data-driven geotechnical engineering applications" aims to contribute to this interdisciplinary research field by taking a modest step forward. The thesis concentrates on five specific research topics: (1) classic ML for classification, (2) classic ML for regression, (3) supervised learning by convolutional neural network (CNN), (4) unsupervised learning by CNN, and (5) deep learning (DL) with 3D input data. These five research topics cover a wide range of contents, including (1) boulder fall volume range prediction, (2) seismic source localization, (3) rock type classification, (4) low-light rock image enhancement, and (5) 3D point cloud filtering. They collectively contribute to the integration of ML in geoscience.

This dataset serves the abovementioned studies. Specifically, Dataset (Chapter 3) contains the datasets used for boulder fall volume range prediction by machine learning algorithms. Dataset (Chapter 4) contains the datasets used for source localization of acoustic emission events using the support vector machine and iterative methods. (Chapter 5) contains the datasets used for automatic rock type classification using HKUDES_Net, a Python program. (Chapter 6) contains the datasets used for low-light rock image enhancement. (Chapter 7) contains the datasets used for automatic and point-wise noise filtering for 3D point clouds. More details are found in the README.txt file.