Neural Network for Crystal Structure Error Classification
A neural network has been trained to classify crystal structure errors in metal–organic framework (MOF) databases.
According to the study,
machine learning models are only as good as the data they are trained on. The approach detects and classifies structural errors, including proton omissions, charge imbalances, and crystallographic disorder.
This can improve the accuracy of computational predictions used in materials discovery, which rely on such databases. However, concerns are growing over the reliability of the underlying datasets, as large crystal structure databases often contain errors.
Author's summary: Neural network improves crystal structure database accuracy.