语义出版本体模型的构成分析——以SPAR本体和Nature本体为例

Composition Analysis of the Semantic Publishing Ontological Model:Taking SPAR ontologies and Nature Ontologies as Examples

[目的 /意义]构建一个良好的本体可为语义出版提供机器可读的数据架构,支持出版物内容的结构化组织和语义化描述,支持出版工作流的术语规范控制,实现网络范围的增值效应。前瞻性思考其合理构成可为我国语义出版本体的构建实践提供参考。[方法 /过程]选取国际学术出版领域分别具有理论和实践代表意义的SPAR本体和Nature本体,深入阐述其构建原则、逻辑结构和内容构成特点,并从多维度进行对比分析,梳理其优势与不足。[结果 /结论]SPAR本体类目全面、属性丰富,突出语义内容的描述。Nature本体结构简洁,灵活实用,注重与外部资源的关联和映射。基于系统分析,认为语义出版本体的构成主要包括出版物和出版工作两方面信息。其中,对出版物重要构成要素(如参考文献与引用信息、学术贡献等)进行深入的建模描述,可增强本体表示能力,更好地支持语义出版的实现;同时,注意做好顶层设计,以实现本体模型的结构简洁与功能高效。

[ Purpose/significance] A well-constructed semantic publishing ontology can provide a machine-readable data structure, which may support the structured organization and semantic description of publication contents, and sup- port the term authority control of publishing workflow, finally realizing the value-added effect within the web. Considering its logical composition prospectively may serve as a reference for the construction of semantic publishing ontology in China. [ Method/process ] Two representative ontologies, SPAR ontologies and Nature ontologies, were thoroughly analyzed and compared from the dimensions of principles, structures, classes and attributes, as well as their strengths and weaknesses. [ Result/conclusion] SPAR ontologies have comprehensive classes and attributes, characterized in rich semantic descrip- tions of contents. Nature ontologies have a concise structure and practical classes, as well as numerous linked data and mappings with external resources. Based on the systematic analysis, it is concluded that semantic ontologies mainly consist of two parts, namely publication and publishing related information. Among other things, independently modeling and deeply describing important publication components can increase the representation ability of models and better support the realization of semantic publishing. Meanwhile, the top-level design is critical for constructing an ontology with a concise structure and highly efficient functions.