摘要：

【目的】 科学出版物的语义化过程中产生了大量的语义数据,分析这些语义数据的特征可以为语义出版实践提供参考和借鉴。【方法】 通过收集目前常用的典型科学出版物语义数据,多角度分析这些数据集的类型、组成要素、生成流程及其支持的应用系统功能,总结这些语义数据在开发、开放、发布中的共性。【结果】 目前主流的科学出版物语义数据集具有相似的数据开发与组织流程,偏向于自然科学领域,具有不同的组织粒度及应用特点。【结论】 科学出版物语义数据的生产与应用仍面临诸多挑战,需要在数据的规范使用、数据本身的质量以及基于这些数据的服务上予以重视。

关键词: 语义出版, 数据出版, 科学数据, 语义数据, 科学出版物

Abstract:

[Purposes] A large number of semantic data are generated in the semantic process of scientific publications, and the analysis of the characteristics of these semantic data can provide references for the semantic publishing practice. [Methods] The typical semantic publishing data were collected, and their similarity in development, openness, and publishment were analyzed from perspectives of their types, content elements, generation process, and application functions. [Findings] Developments and organization processes of these datasets are similar to each other and most of these datasets come from nature and science disciplines with different characteristics of granularity and applications. [Conclusions] Many challenges in productions and applications of semantic publishing datasets still need to be faced, and more attentions should be given to data usage, data quality,and data-based services.

Key words: Semantic publishing, Data publishing, Scientific data, Semantic data, Scientific publication