GBN Health resilience Knowledge Graph (T3.5)

This knowledge graph dataset presents a comprehensive RDF (Resource Description Framework) representation of infectious disease mitigation strategies in Green Building Neighbourhoods (GBNs). Constructed using the Owlready2 Python library after parsing 370+ research articles through GROBID, the dataset contains over tens of thousands of individual entries semantically categorized into risks (using a PESTLE-based taxonomy), stakeholders, technologies, and interventions. The knowledge graph connects these concepts into a structured network enhanced by vector embeddings for improved search capabilities and integration with language models, creating relationships between various building and neighbourhood-scale considerations pertinent to health resilience in sustainable built environments. It incorporates both automatically extracted information and manually created blueprints that provide actionable mitigation strategies, which can be filtered and prioritized according to local contexts. This structured approach transforms disparate research findings into a navigable resource supporting evidence-based decision-making for pandemic resilience. The knowledge graph serves as a bridge between traditional literature reviews and modern LLM-powered knowledge management, enabling stakeholders to efficiently leverage collective intelligence when addressing the complex challenges of disease transmission in urban environments.本知识图谱数据集以资源描述框架（Resource Description Framework，RDF）形式，全面呈现了绿色建筑社区（Green Building Neighbourhoods，GBNs）内的传染病防控策略。本数据集基于Owlready2 Python库构建，经GROBID工具解析370余篇学术文献后，共收录逾数万条语义条目，并按语义分类为基于PESTLE的风险类别、利益相关方、技术及干预措施四大类。本知识图谱将上述概念整合为结构化网络，辅以向量嵌入（vector embeddings）技术以提升检索性能并实现与大语言模型的集成，搭建起可持续建成环境健康韧性相关的各类建筑及社区尺度考量要素间的关联。 本数据集既包含自动抽取的信息，也涵盖手动生成的可落地防控策略方案，可根据本地实际场景进行筛选与优先级排序。这种结构化方法将零散的研究成果整合为可浏览检索的实用资源，为疫情韧性相关的循证决策提供支撑。本知识图谱充当了传统文献综述与现代大语言模型驱动的知识管理之间的桥梁，助力利益相关方在应对城市环境中疾病传播的复杂挑战时，高效利用集体智慧。