First principles calculations informing machine learning framework and visualization system for rapid and generalized gas response prediction in black phosphorus sensors (IMAGE)
Caption
An integrated framework combining first-principles calculations and machine learning was developed to predict gas-sensing performance. Key descriptors such as adsorption energy, adsorption distance, and Fermi level change were extracted from first-principles calculations and used to train machine learning models including Decision Tree, Random Forest, and Extra Trees for intelligent classification of gas response (“1” for response/“0” for non-response). The models were systematically evaluated using ROC-AUC, confusion matrices, and Pearson correlation, forming a complete methodology from descriptor extraction to performance prediction and providing data-driven support for rational design of gas-sensing materials.
Credit
Journal of Advanced Ceramics, Tsinghua University Press
Usage Restrictions
News organizations may use or redistribute this image, with proper attribution, as part of news coverage of this paper only.
License
Original content