Composition based crystal materials symmetry prediction using machine learning with enhanced descriptors

Geometric information such as the space groups and crystal systems plays an important role in properties of materials. Prediction system group thus has wide applications material property estimation structure prediction. Previous works on experimental X-ray diffraction (XRD) density functional theory (DFT) based determination methods achieved outstanding performance, but they are not applicable for large-scale screening materials compositions. There also machine learning models using Magpie descriptors composition determination, their prediction accuracy only ranges between 0.638 0.907 different kinds crystals. Herein, we report improved model predicting inorganic formula information. Benchmark study a dataset downloaded from Materials Project Database shows that our random forest new descriptor set, achieve significant performance improvements compared with previous work scores ranging 0.712 0.961 terms classification. Our large improvement Trained source code freely available athttps://github.com/Yuxinya/SG_predict