Overview / Usage

Mineral exploration involves the collection and analysis of large and diverse datasets. Machine learning algorithms can assist in the analysis of these datasets and can potentially help discover new mineral exploration targets. This project will utilize multiple supervised and unsupervised machine learning algorithms to help integrate and identify hidden correlations and trends in geophysical, geochemical, petrophysical, remote sensing, structural and lithological datasets.

Methodology / Approach

Thus far, I have applied k-nearest neighbors, regression, and random forest algorithms to petrophysical and geochemical datasets collected from the same regional area in Canada and have found no significant correlations or patterns in the datasets that are geologically reasonable.

Technologies Used

Python, scikit-learn