Semantic Resources are a key component of Open Science infrastructures, supporting interoperability, knowledge integration, and semantic analysis across heterogeneous datasets. By combining Semantic Web technologies, knowledge graphs, ontologies, and AI, they make research data more accessible, reusable, and machine-actionable.
Through standards such as RDF, SPARQL, Linked Data, and FAIR principles, Semantic Resources enable semantic annotation, advanced querying, metadata enrichment, and knowledge graph creation, fostering transparent and collaborative research ecosystems. Services such as Ontogenia and the Semantic Data Store (SDS) contribute to the development of interoperable semantic infrastructures for social sciences and Open Science environments.
In the evolving landscape of digital research infrastructures, new solutions are emerging that combine artificial intelligence, Semantic Web technologies, and Open Science principles. Among them, Ontogenia and the Semantic Data Store (SDS) stand out as strategic tools for the creation, management, and enhancement of interoperable semantic knowledge.
Ontogenia: ontology engineering powered by Large Language Models
Ontogenia is a next-generation service designed to support the assisted construction of ontologies and knowledge graphs through the use of Large Language Models (LLMs) and advanced deep learning techniques. Its main goal is to simplify ontology engineering processes by automating complex tasks such as competency question generation, ontology creation, and knowledge graph validation.
One of the project’s most innovative aspects is the use of “metacognitive prompting”, an approach that encourages the language model to reason about its intermediate steps, progressively improving the quality and consistency of the generated knowledge. Through this methodology, Ontogenia enables researchers to transform natural language descriptions, user stories, or research requirements into formal semantic structures that are interoperable and compliant with FAIR principles.
From a methodological perspective, the service builds upon established Semantic Web practices, integrating the eXtreme Design (XD) methodology and the reuse of Ontology Design Patterns (ODPs), which are essential for ensuring modularity, reusability, and consistency in knowledge modelling.
A strategic resource for FOSSR
Within the FOSSR ecosystem, Ontogenia plays a particularly strategic role by supporting:
- the production of FAIR semantic metadata;
- interoperability among heterogeneous datasets;
- semantic documentation of research resources;
- the development of knowledge graphs for social sciences;
- the integration of Open Science and generative AI.
The service is intended for researchers, public and private institutions, and non-profit organisations interested in developing open, interoperable, and semantically enriched data ecosystems.
Semantic Data Store: the backbone of knowledge graphs
Alongside Ontogenia, the Semantic Data Store (SDS) provides the technological infrastructure for storing, managing, and querying semantic data in RDF format through SPARQL. It is a specialised graph database designed to support the creation and exploitation of interoperable knowledge graphs aligned with Semantic Web standards.
The Semantic Data Store enables:
- integration of heterogeneous data sources;
- organisation of complex semantic relationships;
- advanced querying through SPARQL;
- interoperability and data reuse;
- support for Open Science and data-driven research initiatives.
The infrastructure adheres to FAIR principles, ensuring that data are findable, accessible, interoperable, and reusable, while serving as a robust backbone for linked data ecosystems and distributed semantic knowledge environments.
Towards AI-assisted knowledge engineering
The integration of generative AI, ontology engineering, and semantic technologies opens new perspectives for scientific research. Ontogenia and SDS represent an important step towards AI-assisted knowledge engineering platforms, where semantic knowledge construction becomes faster, more scalable, and more accessible, while preserving traceability, explainability, and formal consistency.
Video insights: https://l.cnr.it/99tp9
Further Reading Materials:
https://zenodo.org/records/10793529
https://zenodo.org/records/17466326