Data supporting for: High-Precision Predictive Modeling for BaTiO3–Bi(Mg1/2Ti1/2)O3 Material System Using Automated Machine Learning

An automated machine learning (AutoML) framework integrating Auto-Sklearn and Auto-Keras algorithms was implemented to develop predictive models for the unexplored parameter spaces of the BaTiO3–Bi(Mg1/2Ti1/2)O3 (BT–BMT) material system. The Auto-Sklearn algorithm yielded a high-precision model with an value of 0.9929, significantly surpassing the performance of manually tuned models. This model facilitated the identification of parameter configurations that satisfy X9R temperature stability standards for the dielectric constant across the entire parameter space. Furthermore, the model was employed to prioritize experimental parameters impacting the dielectric properties of the BT-BMT system during the solid-state reaction process. Critical factors identified included sintering temperature, coarse grinding time, post-calcination grinding time, and sintering time. These findings highlight the scalability and adaptability of the AutoML framework, emphasizing its potential to accelerate material discovery and optimization in various domains of materials science.