交叉学科文献数据集构建探讨——以海洋学科为例

Discussion on Construction of Interdisciplinary Literature Data Set: Case Study of Marine Science

[目的/意义]以海洋学科为例,提出一种通过主题、学科结合权威研究机构发文量来选定期刊再构建数据集的方法。[方法/过程]方法先从"全"角度出发,通过主题筛选海洋研究的相关文献,结果显示除传统认为的"海洋"类学科外,有大量的相关文献交叉涵盖在地质、环境等学科分类中,但也存在部分杂质;其次从"准"角度精选,以主题检索文献所在的期刊为基础,利用权威机构在这些期刊中的发文量筛选出132种"涉海"期刊,其结果既包含了海洋类期刊,也包含了综合性及交叉学科期刊;最后以学科分类中的海洋类期刊、主题、机构精选的"涉海"期刊,检索并收录其中的海洋类文献,并作为最终的数据集,经验证数据集中的海洋相关文献率达到93.5%。[结果/结论]本文采用的这种从"涉海"期刊入手的方法既保留了数据库已有的"海洋类"期刊,也合理考虑到海洋分类以外的大量重要"涉海"综合性期刊,在一定程度上解决了交叉学科文献数据集构建中难以"全"而"准"的问题。

[Purpose/significance]The paper takes marine science as an example, and proposes a method to select periodicals for building a literature data set by combining subject and discipline with the number of papers published by authoritative research institu-tions. [Method/process ]The method firstly uses theme screening to obtain marine research literatures for the purpose of “Database In-tegrity”, the results show that in addition to the traditional “Marine” disciplines, there are a large number of related literatures in geol-ogy or environment discipline, but still existing some unrelated periodicals. Secondly, for the purpose of “Database Accuracy”, the method bases numbers of papers published by authoritative research institutions to select 132 journals related to “Sea” from periodicals retrieved by literature theme, these periodicals not only contain marine journals but also contain compre hensive and interdisciplinary journals. Finally, the method is to retrieves and collects marine literatures which are picked on the basis of subject classification of ma-rine journals, research themes, and authoritative research institutions as final data set where correlation rate of marine literature reaches 93.5Z. [Result/conclusion ]This method keeps the original “Marine” periodicals in database, as well as considering plenty of compre-hensive journals related to “Sea”. So, in a certain extent, this method solves problem that it is difficult to achieve “Integrity” and “Ac-curacy” simultaneously, when establishing interdisciplinary literature data set.