Rapid Identification of X-ray Diffraction Spectra Based on Very Limited Data by Interpretable Convolutional Neural Networks

TL;DR

This paper introduces an interpretable CNN trained on limited experimental and theoretical XRD data, achieving rapid and accurate identification of MOFs, with potential applications across various material characterization techniques.

Contribution

The study presents a novel data augmentation method and demonstrates one-to-one material identification using CNNs on limited experimental data.

Findings

Achieved 96.7% top-5 accuracy in MOF identification

Data augmentation with noise and theoretical spectra improves model performance

CNN model reveals interpretability through class activation maps

Abstract

Large volumes of data from material characterizations call for rapid and automatic data analysis to accelerate materials discovery. Herein, we report a convolutional neural network (CNN) that was trained based on theoretic data and very limited experimental data for fast identification of experimental X-ray diffraction (XRD) spectra of metal-organic frameworks (MOFs). To augment the data for training the model, noise was extracted from experimental spectra and shuffled, then merged with the main peaks that were extracted from theoretical spectra to synthesize new spectra. For the first time, one-to-one material identification was achieved. The optimized model showed the highest identification accuracy of 96.7% for the Top 5 ranking among a dataset of 1012 MOFs. Neighborhood components analysis (NCA) on the experimental XRD spectra shows that the spectra from the same material are clustered in groups in the NCA map. Analysis on the class activation maps of the last CNN layer further discloses the mechanism by which the CNN model successfully identifies individual MOFs from the XRD spectra. This CNN model trained by the data-augmentation technique would not only open numerous potential applications for identifying XRD spectra for different materials, but also pave avenues to autonomously analyze data by other characterization tools such as FTIR, Raman, and NMR.