This repository is a culmination of the work related to landslide failure movement identification using a topology-aided ML model. The published work can be found here: https://www.nature.com/articles/s41467-024-46741-7
Please cite our work in case you decide to use our codes: Bhuyan, K., Rana, K., Ferrer, J.V. et al. Landslide topology uncovers failure movements. Nat Commun 15, 2633 (2024). https://doi.org/10.1038/s41467-024-46741-7
"Sample Feature Engineered Data" contains some sample arrays that were generated using TDA. The data can be used to train any ML/DL model. In our study, we utilized the Random Forest model but in theory, any ML or DL model can be used.
We also release the documentation of two inventories from Fan et al. (2019) which are now documented with failure-type information. Please find them in the folder 'Predictions'.