Manufacturing process encoding through natural language processing for prediction of material properties

This dataset encompasses all the data used in the machine learning model presented in the paper titled 'Manufacturing Process Encoding through Natural Language Processing for Prediction of Material Properties,' authored by the same individuals. It consists of historical manufacturing process data and processed information, including both labeled and one-hot encoded data. The paper thoroughly details the hyperparameters employed in the models and provides a comprehensive overview of the training and test sets. Additionally, the dataset includes the detailed code for the neural network structure. Furthermore, it incorporates the code for PCA and K-means analysis as presented in the paper.本数据集涵盖了同一研究团队发表于题为《基于自然语言处理的制造工艺编码以预测材料性能》的论文中所展示的机器学习模型使用的全部数据。本数据集包含历史制造工艺数据与经处理的信息，涵盖带标签数据与独热编码（one-hot encoding）数据两类。该论文详细阐述了模型所采用的超参数（hyperparameter），并对训练集与测试集进行了全面概述。此外，本数据集还包含了神经网络结构的详细代码。更进一步，本数据集收录了该论文中提及的主成分分析（Principal Component Analysis, PCA）与K均值聚类（K-means clustering）分析代码。