Activities & Results

SSH Knowledge Graph

A cornerstone outcome of the project is the creation of an advanced Knowledge Graph for SSH that combines established technologies, offering a hybrid architecture that integrates data from various sources, enhancing data findability, understanding and decision-making.

Deliverables:

The GRAPHIA Technical Architecture sets out the technical foundations for developing and operating the GRAPHIA SSH Knowledge Graph and Data Acquisition Platform. This deliverable consolidates key design decisions and describes the semantic, architectural, and operational layers that together form a coherent, FAIR-aligned, and sustainable infrastructure. The document introduces GRAPHIA’s federated knowledge graph model, which connects autonomous SSH knowledge graphs through a shared interoperability framework rather than centralised ingestion. 

The GRAPHIA Ontology describes the development of the GRAPHIA common data model. In the first section, we highlight the technical choices that drive this work, i.e. the technical architecture, described in the technical architecture and a state of the art on ontology development methodology and existing interoperability frameworks. The second section describes the scope and conceptual design of the GRAPHIA Ontology based on the choice made from the state-of-the-art analysis. The third section describes our methodology to develop the GRAPHIA Ontology, i.e. requirement collection, testing methods, publication and governance of the ontology. The fourth and final section provides an overview of the results of the development work, including an analysis of the initial data sources to be interconnected and of the potential additional candidate sources.

SSH Citations Index

A software framework for the automatic extraction, indexation, enrichment and use of citation data from any SSH resource, increasing the visibility and tracking of research inputs and outputs. This systematic approach consolidates resources for citation data processing across all SSH domains and fosters the reuse of project outcomes in both public and private publishing.

Deliverables:

The Report on Existing Datasets, Methods and Tools for the SSH Citation Index presents the main results of GRAPHIA ’s work on understanding, evaluating, and preparing the technical foundations for the SSH Citation Index modules. Its central contribution is to show, through systematic analysis and empirical evaluation, that bibliographic reference processing in the Social Sciences and Humanities (SSH) raises challenges that are not adequately addressed by approaches developed primarily for STEM disciplines, and to identify concrete ways in which GRAPHIA can respond to these challenges.

LLM4SSH

A Large Language Model tailored for SSH that enhances SSH research by providing domain-specific insights and analytics while offering high performance at lower costs while being able to be deployed on private servers for increased data privacy.

Prototypes for next-generation technological instruments for SSH-> State-of-the-art prototypes that incorporate advanced data acquisition and processing technologies, enable the creation of interactive digital twins and facilitate in-depth and nuanced analyses of SSH cultural objects and digital data. Upgraded instruments will significantly enhance research capabilities and promote interdisciplinary.

Deliverables:

As part of ongoing technical development, the GRAPHIA consortium has documented the deployment of large language model (LLM) services that form a key part of the project’s growing SSH-oriented AI ecosystem. The current infrastructure brings together production-ready and validated LLM services deployed across both cloud-native (OKD) and high-performance computing (HPC) environments, creating a robust foundation for future AI-powered services within GRAPHIA. As the project continues to evolve, both the underlying models and the range of services built upon them will continue to expand in response to technological developments and the needs of the wider GRAPHIA community.

Technical Interoperability Framework: Toolkits, Workflows, APIs

A structured set of guidelines and policies to ensure seamless integration and consistent communication across systems, tools and services in the SSH domain and beyond. It promotes consistency, compatibility and the FAIR principles through standardised protocols and interfaces, enhancing transparency and trust and creating a unified data ecosystem. The robust framework further ensures future exploitation in not only SSH but also other scientific domains to enable and enhance cross-domain research activities.

Deliverables:

With the goal of expanding and strengthening GRAPHIA’s Technical Interoperability Framework, a draft catalogue of innovative workflows was designed to support the connection of heterogeneous and legacy SSH and Cultural Heritage data to the SSH Knowledge Graph. The workflows provide a high-level, FAIR-by-design approach to documenting provenance, data transformations, and semantic readiness. The draft catalogue is openly published in the GRAPHIA Zenodo community and serves as a framing resource for future Knowledge Graph-compliant ingestion pipelines.

The deliverable Catalogue of Next-Generation SSH Instruments and Tools documents a set of advanced hardware and software tools contributed by project partners and relevant to the Social Sciences and Humanities (SSH) data lifecycle, from acquisition and processing to enrichment, interpretation, and publication. The instruments and tools described in this catalogue support the generation and management of multimodal data and contribute to the broader GRAPHIA objective of enabling semantically enriched, interoperable, and graph-ready SSH data (i.e. structured and mapped to support ingestion into the GRAPHIA SSH Knowledge Graph).

Real-World Use Case Implementations

A series of concrete use cases covering both academic and commercial applications that have been defined and included in the project to demonstrate the applicability of project results in the real world. Each use case has been identified and selected for its exploitation potential and will include individual exploitation plans.

Deliverables:

This Industry Hub contains GRAPHIA use cases from commercial, private as well as public organisations working with or providing services to commercial organisation. It showcases how cutting-edge knowledge graph technologies can transform real-world products, services and workflows across sectors. It highlights practical applications where open, interoperable SSH data infrastructure adds value for organisations: from innovative SMEs, publishers, data providers to established tech providers. This hub connects research expertise with commercial needs, accelerating adoption of knowledge graph-enabled services in diverse markets.

More use cases (commercial, academic and others) will be delivered later in the project, an initial presentation of potential use cases can be found in the recording from the webinar: Why a Knowledge Graph for SSH? Concrete Use Cases with GRAPHIA.

Question Answering Over Graphs (Quagga)

Quagga is GRAPHIA’s ecosystem of data and tools for LLM-based knowledge graph question answering. It comprises three components: the Quagga benchmark, a community-crowdsourced evaluation resource pairing natural-language questions with SPARQL queries across SSH knowledge graphs; the Quagga agent, an agentic system built on open-weight LLMs that gives conversational access to those graphs through retrieval-augmented generation; and Explore Your Graphs Engine (EXYGEN) , an API that generates the knowledge-graph metadata grounding query generation from any SPARQL endpoint. Its primary application is a natural-language interface to GRAPHIA’s federated Knowledge Graph and the GoTriple knowledge graph, lowering the technical barrier to exploring them.
ans.

Deliverables: https://quagga.graphia-ssh.eu/home