Machine learning prediction of DFT-derived electrical conductivity in perovskite and double perovskite materials using physically motivated descriptors
- 1Department of Theoretical and Nano Physics, Faculty of Physics, Alzahra University
Abstract
Predicting electrical conductivity in perovskite and double-perovskite materials remains challenging as this property depends on various electronic, chemical, and structural factors. In this work, we evaluate classical machine-learning models trained on different descriptor sets including DFT band gap, non-orbital compositional descriptors, orbital-related descriptors, SOAP structural fingerprints, and a reduced mixed descriptor set to predict DFT-derived transport conductivity. The band gap provides a strong baseline but is insufficient to fully predict the target. The best overall performance is obtained using non-orbital compositional descriptors with Random Forest regression, while orbital-related descriptors achieve nearly comparable accuracy, confirming the importance of valence-electron characteristics. A compact mixed descriptor set preserves nearly the full predictive power of the larger descriptor spaces, showing that accurate prediction can be achieved using a small number of physically motivated variables. In contrast, SISSO showed much lower accuracy, suggesting that sparse symbolic expressions are insufficient to capture the nonlinear relationships underlying our target values. These results demonstrate that physically informed classical machine-learning models can provide an effective surrogate framework for reproducing DFT/BoltzTraP-derived conductivity trends in perovskite and double-perovskite materials.
Related
- Impact of Ni and Mo Substitution on Oxygen Permeability and Electrical Conductivity in BaCo0.4Fe0.4Y0.2O3-δ Perovskite Oxides
- Emergent coherent-incoherent phonon crossover enables ultralow lattice thermal conductivity in 3D double perovskite
- Nonlinear Electrical Conductivity and Thermal Conductivity of g-C3N4/Liquid Silicone Rubber Field Grading Composites
- Enhancing ionic conductivity and interfacial stability of perovskite-type LLTO solid electrolytes via synergistic A-site Sr substitution and polymer layer engineering for safer lithium-ion batteries.
- A Machine Learning Approach to logBB Prediction Combining Molecular Docking and Quantum-Chemical Descriptors
BibTeX
@article{dezashibi2026machine,
title = {{Machine learning prediction of DFT-derived electrical conductivity in perovskite and double perovskite materials using physically motivated descriptors}},
author = {Fatemeh Mohammad Dezashibi and Farinaz Roshani},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-62107-z},
url = {https://doi.org/10.1038/s41598-026-62107-z},
note = {CC BY},
}
RIS
TY - JOUR
TI - Machine learning prediction of DFT-derived electrical conductivity in perovskite and double perovskite materials using physically motivated descriptors
AU - Fatemeh Mohammad Dezashibi
AU - Farinaz Roshani
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-62107-z
UR - https://doi.org/10.1038/s41598-026-62107-z
AB - Predicting electrical conductivity in perovskite and double-perovskite materials remains challenging as this property depends on various electronic, chemical, and structural factors. In this work, we evaluate classical machine-learning models trained on different descriptor sets including DFT band gap, non-orbital compositional descriptors, orbital-related descriptors, SOAP structural fingerprints, and a reduced mixed descriptor set to predict DFT-derived transport conductivity. The band gap provides a strong baseline but is insufficient to fully predict the target. The best overall performance is obtained using non-orbital compositional descriptors with Random Forest regression, while orbital-related descriptors achieve nearly comparable accuracy, confirming the importance of valence-electron characteristics. A compact mixed descriptor set preserves nearly the full predictive power of the larger descriptor spaces, showing that accurate prediction can be achieved using a small number of physically motivated variables. In contrast, SISSO showed much lower accuracy, suggesting that sparse symbolic expressions are insufficient to capture the nonlinear relationships underlying our target values. These results demonstrate that physically informed classical machine-learning models can provide an effective surrogate framework for reproducing DFT/BoltzTraP-derived conductivity trends in perovskite and double-perovskite materials.
SN - 2045-2322
ER -