Discriminating tectonic setting of igneous rocks using biotite major element chemistry-a machine learning approach

Data Set S1 presents the complete working dataset of biotite composition. This dataset has been used to build machine learning (ML) classifiers, i.e., XGBoost and LightGBM models to discriminate among biotite from the five tectonic settings: continental arc, island arc, continental intraplate, oceanic intraplate and rift setting. Sample description, their respective geographic locations, tectonic settings, host rock types, primary or secondary origin, major oxides data (in weight%), atoms per formula unit (apfu) values of major elements, including the list of citation from which samples data are collected, are summarized here.

Data Set S2 contains biotite analyses (major oxide wt.% and atoms per unit formula values) from Neoproterozoic Malani Igneous Suite, Erinpura granite and their equivalent rocks from the Aravalli-Delhi Belt, Rajasthan, northwestern India (including list of citations from which data are collected). This dataset has been used to constrain the Neoproterozoic geodynamic setting of the Aravalli-Delhi Belt, northwestern India, by applying previously built ML models.  Additionally, the tectonic setting for each sample (classified by the ML models) is mentioned here.