From Spectra to Geography: Intelligent Mapping of RRUFF Mineral Data

TL;DR

This paper presents a machine learning approach using ConvNeXt1D neural networks to accurately classify mineral spectra from the RRUFF database by country, achieving 93% accuracy and enabling geolocation of mineral samples.

Contribution

Introduces a novel ConvNeXt1D-based framework for geolocating minerals from spectral data, leveraging a large dataset for high-accuracy classification.

Findings

Achieved 93% average classification accuracy.

Successfully classified over 32,900 samples from 101 countries.

Demonstrated the effectiveness of spectral data in geolocation tasks.

Abstract

Accurately determining the geographic origin of mineral samples is pivotal for applications in geology, mineralogy, and material science. Leveraging the comprehensive Raman spectral data from the RRUFF database, this study introduces a novel machine learning framework aimed at geolocating mineral specimens at the country level. We employ a one-dimensional ConvNeXt1D neural network architecture to classify mineral spectra based solely on their spectral signatures. The processed dataset comprises over 32,900 mineral samples, predominantly natural, spanning 101 countries. Through five-fold cross-validation, the ConvNeXt1D model achieved an impressive average classification accuracy of 93%, demonstrating its efficacy in capturing geospatial patterns inherent in Raman spectra.