Spatial data mining with machine learning to reveal mineral exploration targets under cover in the Gawler Craton, South Australia. The pre-print manuscript is availble in the "GAWLER" directory.
Use Mines and Minerals dataset to teach the model values from the datasets that are associated with mines and known deposit locations. Then apply that trained algorithm on the entire area.
The dockerfile will build an environment which has all the required packages.
Use results-sa.ipynb for exploring individual commodities
Use results-sa-test1.py to generate the first part of the target test set in the Gawler region.
Use results-sa-test2.py to do the second part of the target test set generation.
Regularized input data is in SA-DATA folder.
Model output (exploration targeting maps) are in MAPS.
Model training and target datasets generated in ML-DATA.
References and links to original data licenses can be found in the manuscript.