Shanghai Materials Genome Institute, Shanghai University, 200444, Shanghai, P. R. China.
School of Physics and Electronic Science, East China Normal University, 200241, Shanghai, P. R. China.
Chem Asian J. 2022 Nov 16;17(22):e202200771. doi: 10.1002/asia.202200771. Epub 2022 Sep 26.
New ternary gold alloys with low resistivities (ρ) were screened out via an interpretable machine learning strategy by using the support vector regression (SVR) model integrated with SHAP analysis. The correlation coefficient (R) and the root mean square error (RMSE) of test set were 0.876 and 0.302, respectively, indicating the strong generalization ability of the model. The average ρ of top 10 candidates was 1.22×10 Ω m, which was 41% lower than the known minimum of 2.08×10 Ω m. The outputs of SVR model were analyzed with the critical SHAP values including first ionization energy of C-site (584 kJ ⋅ mol ), electronegativity of C-site (1.72) and the second ionization energy of B-site (1135 kJ ⋅ mol ), respectively. Moreover, an online web server was developed to share the model at http://materials-data-mining.com/onlineservers/wxdaualloy.
通过支持向量回归（SVR）模型与 SHAP 分析集成的可解释机器学习策略，筛选出电阻率（ρ）低的新型三元金合金。测试集的相关系数（R）和均方根误差（RMSE）分别为 0.876 和 0.302，表明该模型具有很强的泛化能力。前 10 名候选物的平均 ρ 值为 1.22×10Ω·m，比已知的最小值 2.08×10Ω·m 低 41%。使用关键的 SHAP 值（包括 C 位的第一电离能（584kJ·mol）、C 位的电负性（1.72）和 B 位的第二电离能（1135kJ·mol））分析 SVR 模型的输出。此外，还开发了一个在线网络服务器，可在 http://materials-data-mining.com/onlineservers/wxdaualloy 上共享该模型。
