ISBN:

URL:

Abstract:

<p>This paper employed Random Forests (RF) to generate several Mineral Prospectivity maps for orogenic gold in the Geraldton area, located within the Wabigoon Tectonic subprovince of Ontario, Canada. Various issues pertinent to the Mineral Prospectivity mapping process are presented and proposed solutions to these key challenges are suggested. Additionally, multiple methods are proposed to analyze text-based geoscientific information derived from geological maps, including a novel application of Natural Language Processing (NLP) to delineate the sources and traps of gold mineral systems. The Mineral Prospectivity maps generated have located new possible areas for gold exploration. Concerning the key issues addressed in the paper, (1) the results from NLP have contributed to significant predictor maps for gold exploration, (2) the method for creating a non-deposit class for input to the random forests machine learning algorithm was found to involve creating points at least 2&nbsp;km from existing Au deposits\occurrences, (3) a weighting method for existing Au deposits based on tonnage produced was successfully introduced and (4) methods of producing ensemble combinations of the Mineral Prospectivity maps were produced and discussed. The results produced from the paper should significantly enhance Au exploration in the Geraldton area of Ontario, Canada.</p>