摘要
本文在分析科技资源特点的基础上,对国内外科技资源元数据标准的研究现状进行了梳理和总结。通过梳理发现,在目前的研究工作中,普遍缺乏对用户需求的考虑。在此基础上,本文提出了一种基于用户需求挖掘的研究视角,借助目前蓬勃发展的自然语言处理和机器学习等技术,充分考虑元数据服务终端用户的真实需求,实现元数据域的自动生成和热度排序,以此提高元数据服务和应用的效率,进而促进科技资源共享的进一步发展。
By investigating the data of current science and technology resources, the paper makes a thorough analysis of metadata standards for science and technology in China and abroad, finding out that current researches lack considerations on the real requirements of end users. To solve this, the paper proposes a new way to fulfill related researches, thus to draw support from modern natural language processing and machine learning approaches. The new method is designed to generate metadata domains and their ranks based on the history requirements of end users, so to improve the efficiency of metadata services, and help end users share the science and technology resources.
作者
赵启阳
张辉
王志强
ZHAO Qi-yang;ZHANG Hui;WANG Zhi-qiang(Beihang University,School of Computer Science and Engineering;National Engineering Research Center for Science and Technology Resources Sharing Service;China National Institute of Standardization)
出处
《标准科学》
2019年第3期12-17,共6页
Standard Science
基金
国家重点研发计划项目"分布式科技资源体系及服务评价技术研究"(编号为2017YFB1400200)资助
关键词
元数据
科技资源
自然语言处理
机器学习
metadata
science and technology resources
natural language processing
machine learning