According to the paper’s authors, their main objective was to improve traditional geochemical indicators plagued by high false-positive rates.
To achieve such a goal, the researchers developed two algorithms, which they called ‘random forest’ and ‘deep neural network.’ They formulated the models using a global dataset of zircon chemistry, which is normally employed to evaluate the porphyry copper deposits in magma.
In detail, they focused the models on 15 trace elements. They then validated the models with independent data sets from two well-characterized porphyry copper deposits in south-central British Columbia, Canada, and Tibet, China.
Both models resulted in a classification accuracy of 90% or greater. The ‘random forest’ model exhibited a false-positive rate of 10%, whereas the ‘deep neural network’ model had a 15% false-positive rate. In comparison, traditional metrics report false positives at a…
