摘要
由一名新闻记者-机器人与机器学习每日新闻的工作人员新闻编辑-调查人员发布了关于人工智能的新报告。根据中国人民代表大会上海的新闻报道，NewsRx编辑称，“原材料和多步纯化工艺的复杂性，为高纯金属的生产建立普遍适用的工艺参数提出了相当大的技术挑战。”我们的新闻记者从上海大学的研究中得到一句话：“机器学习已经成为材料科学领域不可缺少的工具，它有助于准确预测目标变量，加速过程优化，”本研究探讨了利用高精度机器学习模型预测垂直区域精炼后高纯铟中残余杂质含量的方法,利用82个实验数据集,通过贝叶斯优化确定了XGBoost和LightGBM模型的最优超参数,XGBoost和LightGBM模型的平均绝对误差为(MAEs)。0.022和0.023分别由留一交叉验证确定(LOOCV)。它们与先前建立的岭回归模型(MAE=0.024)的预测性能相当，促使人们探索融合技术，包括均值、加权和叠加FUS离子，以进一步提高准确性。值得注意的是，加权融合模型显示出最优的预测能力，并得到综合评估指标的支持，包括MAE为0.020，均方根误差(RMSE)为0.026，决定系数（2分）为0.830."
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – Investigators publish new report on ar tificial intelligence. According to news reporting out of Shanghai, People’s Rep ublic of China, by NewsRx editors, research stated, “The complexity of raw mater ials and multi-step purification processes presents considerable technical chall enges in establishing universally applicable process parameters for the producti on of high-purity metals.” Our news journalists obtained a quote from the research from Shanghai University : “Machine learning has emerged as an indispensable tool in the field of materia ls science, facilitating the accurate prediction of target variables and acceler ating process optimization, thereby yielding substantial reductions in both expe rimental costs and time. This study explores the utilization of high-precision m achine learning models to predict the residual impurity content in high-purity i ndium after vertical zone refining. A dataset comprising 82 experimental dataset s was employed to determine the optimal hyperparameters for XGBoost and LightGBM models through Bayesian optimization. The XGBoost and LightGBM models demonstra ted mean absolute errors (MAEs) of 0.022 and 0.023, respectively, as determined via leave-oneout cross-validation (LOOCV). Their comparable predictive performa nce to the previously established Ridge regression model (MAE = 0.024) prompted the exploration of fusion techniques, including mean, weighted, and stacking fus ion, to further enhance accuracy. Remarkably, the weighted fusion model exhibite d the most optimal predictive capabilities, supported by comprehensive evaluatio n metrics, including an MAE of 0.020, root mean squared error (RMSE) of 0.026, a nd a coefficient of determination (R2 score) of 0.830.”
NETL
NSTL