用新版LODE辅助理解科学知识图谱的本体
Supporting the understanding of ontologies for scientific knowledge graphs with the new version of LODE
AI总结：
本文提出新版可扩展框架LODE，用于生成语义制品的人类可读文档，通过对SKG-O本体的文档生成，提升了模块化科学知识图谱模型的可访问性与复用性。
AI中文摘要：
科学知识图谱（SKGs）依赖日益复杂且异构的语义模型，其复用与互操作性要求本体及其他语义制品既能被机器理解也能被人类理解。然而，现有的本体文档工具对现代面向复用的建模实践支持有限，这类实践中实体分布在导入模块间并通过跨模块注释丰富。本文提出新版LODE，这是一个经重新设计的可扩展框架，用于生成语义制品的人类可读文档。LODE保留了其前身确立的单页、W3C推荐标准风格的文档范式，还引入了模块化架构，将制品解释（Reader）、内部表示（Model）与渲染（Viewer）分离。作为Web服务，LODE支持OWL本体的文档生成，并具备额外文档功能，如实体独立文档、RDF实体 provenance、Markdown渲染。我们通过对科学知识图谱互操作性框架本体（SKG-O）的文档生成演示了LODE的适用性，展示其如何提升模块化科学知识图谱模型的可访问性与复用性。
英文摘要：
Scientific Knowledge Graphs (SKGs) rely on increasingly complex and heterogeneous semantic models, whose reuse and interoperability require ontologies and other semantic artefacts to be understandable by both machines and humans. However, existing ontology documentation tools provide limited support for modern reuse-oriented modelling practices, where entities are distributed across imported modules and enriched through cross-module annotations. This paper presents the new version of LODE, a re-engineered and extensible framework for generating human-readable documentation of semantic artefacts. LODE preserves the established single-page, W3C Recommendation-style documentation paradigm of its legacy predecessor, and introduces a modular architecture that separates artefact interpretation (Reader), internal representation (Model), and rendering (Viewer). Served as a web service, LODE supports the documentation of OWL ontologies and enables additional documentation features (e.g., entity stand-alone documentation, RDF entity provenance, Markdown rendering). We demonstrate the applicability of LODE through the documentation of the Scientific Knowledge Graph Interoperability Framework ontology (SKG-O), showing how it improves the accessibility and reuse of modular scientific knowledge graph models.
