SuperMat

SuperMat是由日本国家材料科学研究所（NIMS）和法国国家信息与自动化研究所（Inria）合作创建的一个超导材料相关科学出版物的链接注释数据集。该数据集包含142篇科学文章，总计16052个实体和1398个链接，这些链接被分为六个类别：材料名称、类别和属性；与超导临界温度（Tc）相关的链接；以及应用压力或测量方法等参数条件。SuperMat的构建是通过计算机科学家和材料科学家的合作实现的，其高质量通过领域专家的验证得到保证。数据集的应用领域包括材料信息学、机器学习、数据集、超导体和文本与数据挖掘，旨在解决超导材料科学领域中数据驱动材料设计的问题。

SuperMat is a linked annotated dataset for superconducting material-related scientific publications, developed through the collaboration between the National Institute for Materials Science (NIMS, Japan) and the National Institute for Research in Computer Science and Automation (Inria, France). This dataset includes 142 scientific articles, with a total of 16,052 entities and 1,398 links, which are divided into six categories: material names, categories and attributes; links related to superconducting critical temperature (Tc); and parametric conditions such as applied pressure or measurement methods. SuperMat was built through the joint efforts of computer scientists and materials scientists, and its high quality is ensured by validation from domain experts. The application fields of this dataset cover materials informatics, machine learning, datasets, superconductors, and text and data mining, with the goal of solving the problems of data-driven materials design in the field of superconducting materials science.