The Dataset of the article “Machine Learning-Assisted Design of High-Temperature BSPT-Based Piezoelectric Ce-ramics with Enhanced Dual Properties”

This dataset contains 200 BSPT (BiScO3–PbTiO3)-based piezoelectric ceramic samples collected from peer-reviewed published literature. Each entry includes composition information, elemental descriptors, Curie temperature (TC), piezoelectric coefficient (d33), and corresponding literature references with DOIs to ensure full traceability.This dataset supports reproducible research in data-driven design of high-temperature piezoelectric ceramics and facilitates further development of machine learning approaches for functional ceramic materials.