CALLISTO Knowledge Graph

The CALLISTO knowledge graph represents various domains of knowledge addressed within the EU-funded CALLISTO project. CALLISTO aims to bridge the gap between Copernicus Data and Data and Information Access Service (DIAS) providers and users in different domains by providing AI solutions that effectively add value to large amounts of satellite data. The project focuses on the Earth Observations (EO) domain and its relationship with four Pilot Use Cases (PUCs): 1) Common Agricultural Policy (CAP) monitoring, 2) water quality assessment, 3) air quality assessment and journalism, and 4) land border surveillance. Each PUC addresses a particular domain and contributes multiple datasets to the project, which come in diverse formats (e.g., XML, JSON, CSV). The CALLISTO ontology has been developed to handle this multi-domain data by offering a semantic representation of each domain and the connections between them. The knowledge graph was then generated by mapping the PUCs datasets to the ontology using RDF Mapping Language (RML). Knowledge graphs aid in linking data from various domains, generating new knowledge, and inspecting recurring patterns that can be used in simulation and prediction models (i.e., using artificial intelligence and deep learning algorithms). Another benefit of employing KGs is the ability to link them to other Linked Open Data (LOD) for data integration and analytics. Applications such as geospatial question answering, geospatial data retrieval, and cross-domain semantic data-driven applications could use it as an underlying data source. The CALLISTO knowledge graph is available in an RDF format; it is associated with the CALLISTO ontology. For further details about the ontology, please refer to deliverable D6.1: The CALLISTO ontologies and semantic indexing. The ontology and knowledge graph were developed by the Fraunhofer Institute for Intelligent Analysis and Information Systems (IAIS). In collaboration with PUC members, the ontology and knowledge graph were developed and evaluated.CALLISTO知识图谱（CALLISTO knowledge graph）涵盖了欧盟资助的CALLISTO项目所涉及的各领域知识。该项目旨在弥合哥白尼数据（Copernicus Data）与不同领域的数据与信息访问服务（Data and Information Access Service，DIAS）提供商及用户之间的鸿沟，通过人工智能解决方案为海量卫星数据有效赋能增值。项目聚焦地球观测（Earth Observations，EO）领域及其与四个试点用例（Pilot Use Cases，PUCs）的关联：1）共同农业政策（Common Agricultural Policy，CAP）监测、2）水质评估、3）空气质量评估与新闻报道、4）陆地边境监视。每个试点用例对应特定领域，并为项目贡献多份格式多样的数据集（如XML、JSON、CSV）。 为处理这类多领域数据，研究团队开发了CALLISTO本体（CALLISTO ontology），用于为各领域及其间关联提供语义表示。随后，借助RDF映射语言（RDF Mapping Language，RML）将各试点用例的数据集映射至该本体，从而生成本知识图谱。 知识图谱有助于关联跨领域数据、生成新知识，以及挖掘可用于模拟与预测模型（即借助人工智能与深度学习算法）的重复模式。采用关联开放数据（Linked Open Data，LOD）的另一项优势在于，可将其与其他关联开放数据进行关联，以实现数据集成与分析。地理空间问答、地理空间数据检索以及跨领域语义数据驱动应用等场景，均可将该知识图谱作为底层数据源。 CALLISTO知识图谱以RDF格式提供，并与CALLISTO本体绑定。如需了解该本体的更多细节，请参阅交付件D6.1：《CALLISTO本体与语义索引》。该本体与知识图谱由弗劳恩霍夫智能分析与信息系统研究所（Fraunhofer Institute for Intelligent Analysis and Information Systems，IAIS）开发，并联合各试点用例成员完成了开发与评估工作。