Evidence-based recommender system for high-entropy alloys

Abstract Existing data-driven approaches for exploring high-entropy alloys (HEAs) face three challenges: numerous element-combination candidates, designing appropriate descriptors, and limited biased existing data. To overcome these issues, here we show the development of an evidence-based material recommender system (ERS) that adopts Dempster–Shafer theory, a general framework reasoning with uncertainty. Herein, without using model, collect combine pieces evidence from data about HEA phase existence alloys. evaluate ERS, compared its HEA-recommendation capability those matrix-factorization- supervised-learning-based systems on four widely known datasets up-to-five-component The k -fold cross-validation suggests ERS outperforms all competitors. Furthermore, shows good extrapolation capabilities in recommending quaternary quinary HEAs. We experimentally validated most strongly recommended Fe–Co-based magnetic (namely, FeCoMnNi) confirmed thin film body-centered cubic structure.