Machine learning based models for estimating seismically induced slope displacement for subduction zone earthquakes

Estimating the seismic performance of a slope system in subduction earthquakes zone is crucial in performance-based geotechnical earthquake engineering. This project includes Python scripts and Jupyter notebooks that implement the machine-learning-based slope displacement models developed by Macedo and Liu (2022) for subduction zone earthquakes. The implemented machine learning models include ridge regression, support vector regression, random forest, gradient boosting decision tree, and artificial neural network models. These models take as input the yield coefficient (ky), earthquake magnitude (M), fundamental period (Ts), peak ground velocity (PGV), and spectral acceleration at 1.3Ts (Sa(1.3Ts)) and output the median and standard deviation of the slope displacement. The details on how to use the scripts and files can be found in the readme file.

References:
Macedo, J. & Liu, C., (2022). Machine learning-based procedures for estimating seismically-induced landslides in subduction tectonic settings. USGS report