Dataset and Code for “A Domain-Informed Automated Machine Learning Framework for Dielectric Ceramic Design: Application to the BaTiO₃–Bi(Mg₁/₂Ti₁/₂)O₃ System”
Published: 31 October 2025| Version 1 | DOI: 10.17632/7gybvssrpf.1
Contributor:
PQ QinDescription
This dataset contains the raw data, feature descriptors, and Python source codes used in the study “A Domain-Informed Automated Machine Learning Framework for Dielectric Ceramic Design: Application to the BaTiO₃–Bi(Mg₁/₂Ti₁/₂)O₃ System”. It includes the curated dielectric property dataset (313 entries), feature-selection scripts, Auto-Sklearn/Auto-Keras model configurations, and the uncertainty quantification module for predicting dielectric constants and X9R compliance. The dataset supports reproducibility and reuse for research on automated machine learning, uncertainty quantification, and dielectric ceramic materials design.
Files
Categories
Materials Science, Artificial Intelligence, Ceramics, Machine Learning