Optimizing Synthesis Parameters of BiFeO₃ Nanoparticles with Bayesian Neural Networks

The research focuses on improving synthesis parameters for BiFeO<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</inf> nanoparticles, a material with substantial promise in many applications. Conventional approaches frequently fail to precisely forecast the most favorable circumstances for synthesis due to the intricate nature of the process. This article introduces a BNN model as a solution to these restrictions. The BNN model underwent training and validation using datasets obtained from Kaggle and the Materials Project. It achieved a training accuracy of 98.5% and a validation accuracy of 97.8%. The model achieved an F1 score of 0.975 and a recall of 0.980, surpassing the performance of current technology in the sector. The findings presented here showcase the efficacy of the suggested model in delivering precise and dependable predictions, rendering it a helpful instrument for academics and practitioners in the realm of material synthesis.