Refined estimation of Li in mica by a machine learning method

Abstract

Li-rich micas are crucial in the exploration for and exploitation of Li resources. The determination of Li in mica using classical bulk chemical methods or in-situ microanalytical techniques is expensive and time-consuming and has a high-quality requirement for micas and reference materials. Although simple linear and nonlinear empirical equations have been proposed, they are inconsistent with the complex physico-chemical mechanisms of Li incorporation and commonly lead to large errors. In this study, we introduce a refined method of multivariate polynomial regression using a machine learning algorithm to estimate Li from multiple major oxide abundances. The performance of our regression model is evaluated using the coefficient of determination (R 2 ) and the root-mean-square error (RMSE) of the independent test sets. The best-performed models show R 2 of 0.95 and a RMSE of 0.35 wt% for the test set of dataset 1 (all compiled data, n = 2124) and R 2 of 0.96 and a RMSE of 0.22 wt% for the This is the peer-reviewed, final accepted version for American Mineralogist, published by the Mineralogical Society of America.The published version is subject to change. Cite as Authors (Year) Title. American Mineralogist, in press.

Machine-learning oxybarometer developed using zircon trace-element chemistry and its applications

“…Recently, several studies have demonstrated that the data-driven machine learning methods can be powerful tools for solving complex problems in mineralogy, petrology, and geochemistry (Petrelli and Perugini 2016;Chen et al 2021;Huang et al 2022;Lin et al 2022;Nathwani et al 2022;Qin et al 2022;Wang et al 2022;Zou et al 2022), and for the construction of thermobarometers (Petrelli et al 2020;Higgins et al 2022;Jorgenson et al 2022;Li and Zhang 2022), without having any a priori knowledge. These research advances suggest that the machine learning method has the potential to be used for calibrating a mineral chemical-based oxybarometer.…”

A Machine Learning Based‐Approach to Predict the Water Content of Mid‐Ocean Ridge Basalts

“…This means that it has the potential to explore relationships between H 2 O and several other elements simultaneously and predict water contents based on these relationships. Recently, more and more studies have utilized ML methods to solve regression problems in mineralogy and petrology (Graw et al., 2021; Huang et al., 2022; Jorgenson et al., 2022; Lösing & Ebbing, 2021; Petrelli et al., 2020; Ptáček et al., 2020; Thomson et al., 2021; L. Wang et al., 2022), which demonstrates their potential to regress water concentration data in MORBs as well. Although a large number of MORB glasses around the world have been analyzed for their water contents, there are still abundant samples which have only been analyzed for major and trace elements, yet provide a good data foundation for ML research.…”