Predicting titanite genetic types from trace elements: a machine learning approach
- 8 January 2026
- journal article
- Published by Mineralogical Society of America in American Mineralogist: Journal of Earth and Planetary Materials
- Vol. 111 (8) , 1259-1274
Abstract
Titanite, a ubiquitous accessory mineral in diverse rock types, provides robust constraints on geochronology, petrogenesis, and mineralization processes. Despite their diagnostic potential, the trace element signatures of titanite remain underexploited in geological interpretations. Moreover, traditional low-dimensional analysis suffers from well-documented inefficiencies and interpretive subjectivity. In this study, after comparing the performance of six models, we implemented an interpretable machine-learning approach (LightGBM) to relate titanite trace-elements to genetic types Our study covers 14,064 sets of titanite trace element data across five different genesis types (hydrothermal, metamorphic, recrystallized, inherited, magmatic), along with 69 sets obtained from the Mesozoic granite in Jiaodong Peninsula. The LightGBM classifier demonstrated high accuracy in classifying titanite trace-element data by genetic type, achieving 96% overall accuracy with a weighted F1 score of 0.9529. An independent validation on 226 metamorphic titanite from Greenland UHP gneisses correctly classified 205 cases. Additionally, LightGBM predicted 66 titanite from the Jiaodong Mesozoic granites to be of magmatic origin, consistent with their petrographic characteristics. SHAP (SHapley Additive exPlanations) analysis of our model reveals that Nb, Hf, Th/U, Lu, CeAnom, Th, Eu, EuAnom, U, Zr, La, and Lu/Hf constitute the most diagnostically significant variables for discriminating titanite genetic types based on trace element composition, and they further reveal the high coupling and complexity among trace elements in genetic discrimination. In addition, we introduce a visualization approach that combines LightGBM with t-Distributed Stochastic Neighbor Embedding (t-SNE) to classify titanite by its genesis, achieving pronounced separation between metamorphic and magmatic titanite in high-dimensional feature space. These findings establish that our approach provides a robust framework for both deciphering titanite origins from trace element compositions and predicting titanite genesis across diverse geological settings worldwide.Keywords
This publication has 69 references indexed in Scilit:
- Machine Learning Prediction of Ore Deposit Genetic Type Using Magnetite GeochemistryNatural Resources Research, 2022
- Machine learning for geochemical exploration: classifying metallogenic fertility in arc magmas and insights into porphyry copper deposit formationMineralium Deposita, 2022
- In situ LA-ICP-MS U–Pb geochronology and trace element analysis of hydrothermal titanite from the giant Zhuxi W (Cu) skarn deposit, South ChinaMineralium Deposita, 2018
- How do granitoid magmas mix with each other? Insights from textures, trace element and Sr–Nd isotopic composition of apatite and titanite from the Matok pluton (South Africa)Contributions to Mineralogy and Petrology, 2017
- LA-ICP-MS mineral chemistry of titanite and the geological implications for exploration of porphyry Cu deposits in the Jinshajiang – Red River alkaline igneous belt, SW ChinaMineralogy and Petrology, 2014
- Redistribution of HFSE elements during rutile replacement by titaniteContributions to Mineralogy and Petrology, 2010
- A thermobarometer for sphene (titanite)Contributions to Mineralogy and Petrology, 2007
- Stability of feature selection algorithms: a study on high-dimensional spacesKnowledge and Information Systems, 2006
- A Decision-Theoretic Generalization of On-Line Learning and an Application to BoostingJournal of Computer and System Sciences, 1997
- Geochemical implications of rare earth element patterns in hydrothermal fluids from mid-ocean ridgesGeochimica et Cosmochimica Acta, 1994