Democratising and making sense out of heterogeneous scholarly content

Información del proyecto

SciLake

Identificador del acuerdo de subvención: 101058573

DOI

10.3030/101058573

Proyecto cerrado

Fecha de la firma de la CE 28 Noviembre 2022

Fecha de inicio 1 Enero 2023

Fecha de finalización 31 Marzo 2026

Financiado con arreglo a

• Research infrastructures

Coste total

€ 4 809 450,00

Aportación de la UE

€ 4 809 449,00

Coordinado por

ATHINA-EREVNITIKO KENTRO KAINOTOMIAS STIS TECHNOLOGIES TIS PLIROFORIAS, TON EPIKOINONION KAI TIS GNOSIS
Greece

Periodic Reporting for period 2 - SciLake (Democratising and making sense out of heterogeneous scholarly content)

Período documentado: 2024-07-01 hasta 2026-03-31

Resumen del contexto y de los objetivos generales del proyecto

SciLake is a Horizon Europe project that leverages Scientific Knowledge Graphs (SKGs) to establish the concept of the Scientific Lake: an interoperable ecosystem for the creation, integration, enrichment, management, and exploitation of scholarly knowledge. The project addresses key challenges faced by the European research ecosystem, including the fragmentation of scholarly information, limited interoperability among research infrastructures, difficulties in assessing research impact and reproducibility, and barriers to cross-disciplinary knowledge discovery.
In alignment with the objectives of the European Open Science Cloud (EOSC), SciLake promotes open, interoperable, and reusable research infrastructures. Its objectives were to: (i) overcome the heterogeneity of scholarly content, (ii) democratise the creation and use of SKGs, (iii) facilitate the identification of research trends and valuable research outputs, (iv) support reproducibility assessment, (v) validate the developed services in real-world pilots, and (vi) contribute to the EOSC ecosystem through standards and reusable assets.
The project delivered a two-tier ecosystem consisting of the Scientific Lake infrastructure and a set of value-added applications for impact-driven scientific knowledge discovery and reproducibility assistance. Its technologies were deployed and validated through five pilot-specific SKGs covering Cancer, Neuroscience, Energy, Transport-CCAM, and Transport-Maritime research, demonstrating their applicability across diverse scientific communities.

Trabajo realizado desde el comienzo del proyecto hasta el final del período abarcado por el informe y los principales resultados hasta la fecha

SciLake designed, developed, integrated, and validated a complete ecosystem for the creation and exploitation of Scientific Knowledge Graphs. A major achievement was the delivery of the Scientific Lake service, a suite of interoperable components supporting data acquisition, knowledge graph creation, graph interlinking, multilingual processing, graph analytics, querying, and unified access to heterogeneous scholarly knowledge.
The project contributed to the Scientific Knowledge Graph Interoperability Framework (SKG-IF) and delivered the Lake API, providing a common interoperability and access layer across multiple scientific knowledge graphs. Significant advances were achieved in graph generation and profiling, entity resolution, graph analytics, large-scale graph processing, impact assessment, topic analysis, research artefact identification, citation-context analysis, and reproducibility assessment.
Building on this infrastructure, SciLake delivered two advanced applications: the Smart Impact-driven Discovery Service and the Smart Reproducibility Assistance Service. These services support the identification of influential research outputs, emerging trends, reproducibility signals, and relationships among publications, datasets, software, and other research artefacts.
The project deployed and evaluated five pilot-specific Scientific Knowledge Graphs covering Cancer, Neuroscience, Energy, Transport-CCAM, and Transport-Maritime research. The outcomes include five reusable SKG datasets, a portfolio of software components and services, the Lake API, contributions to SKG-IF, the two value-added applications, and 19 scientific publications (6 journal articles, 12 conference/workshop papers, and 1 magazine contribution).

Avances que van más allá del estado de la técnica e impacto potencial esperado (incluida la repercusión socioeconómica y las implicaciones sociales más amplias del proyecto hasta la fecha)

SciLake delivered a comprehensive set of scientific, technical, and infrastructure results centred on interoperable Scientific Knowledge Graphs. Key outcomes include the Scientific Lake service, the Smart Impact-driven Discovery Service, the Smart Reproducibility Assistance Service, the Lake API, five domain-specific SKGs, contributions to SKG-IF, and a portfolio of reusable software, datasets, and publications.
The project generated advances in knowledge graph creation, graph analytics, graph query optimisation, entity resolution, large-scale graph processing, impact assessment, and reproducibility analysis. Novel results include field-normalised impact indicators, indicators for datasets and software, topic-evolution analytics, research artefact recognition, citation-context analysis, and reproducibility indicators.
Beyond its technical contributions, SciLake strengthened the open science and EOSC landscape through interoperable datasets, open-source software, reusable services, and contributions to emerging interoperability standards. The project produced 5 SKG datasets and 19 scientific publications.
To maximise long-term impact, further efforts should focus on adoption by research infrastructures, EOSC initiatives, and scientific communities. Key needs include the sustained operation of the developed services and datasets, continued evolution of SKG-IF, integration with EOSC infrastructures, and follow-up initiatives building on the project’s exploitable results. The exploitation strategy developed with support from the Horizon Results Booster identified concrete pathways for sustaining and further developing the project’s key assets.