论文部分内容阅读
【目的】结合机构产出SCI论文统计需求,设计一款自动甄别目标机构作者和实验室的软件。【应用背景】可辅助论文统计部门快速准确识别机构论文作者和实验室(部门),进而获得机构作者和实验室的论文产出分布情况。【方法】从技术上实现综合利用相同研究单元内作者合作较多的科研特点、自定义作者唯一关键词或合作者字段以及SCI数据库作者相关字段的文本特征来甄别目标机构作者。【结果】允许用户通过目标机构人员名单维护来实现SCI论文作者甄别的自动化和高准确度。【结论】有效解决SCI论文中文作者因拼音写法多样且易重名而造成作者相关论文数据难以准确统计的问题,其设计思路也适用于EI及其他数据库论文作者甄别。
【Objective】 According to the statistical demand of SCI papers produced by institutions, a software to automatically identify the authors and laboratories of target institutes is designed. 【Background】 Assist dissertation statistical department to quickly and accurately identify institutional dissertation authors and laboratories (departments), and then obtain the dissertation outputs of institutional authors and laboratories. 【Method】 From the technical characteristics of the research on the comprehensive utilization of the author’s cooperation in the same research unit, the author identifies the target authors by defining the author’s unique keywords or collaborators ’fields and the textual features of the authors’ relevant fields in the SCI database. 【Results】 Allows the user to achieve the automation and high accuracy of screening of SCI dissertation authors through the maintenance of the list of target agencies. 【Conclusions】 This paper effectively solves the problem that Chinese author of SCI dissertation is difficult to get accurate statistical data due to the diversity of Pinyin writing and easily renamed. The design idea also applies to EI and other authors of database dissertations.