Best Practices

The African Development Bank is a leader in documentation of development project activities, and of policies that promote the disclosure and access to project activities documentation. This paper discusses what a business classification scheme (BCS) is and how BCS semantics and metadata support risk and compliance use cases that are common to multilateral development organizations as well as other public and commercial organizations with public accountability. Best practices in the design of a BCS taxonomy with metadata inherited from each hierarchical level and methods for automating metadata population are briefly described.

At ZBW the efforts to automate the subject indexing process have started as early as 2000, and since 2014 the necessary applied research in the field of artificial intelligence is done in-house. However, the prototypical machine learning solutions that the researchers had developed until 2019 were yet to be integrated into productive operations at the library. Therefore in 2020 a pilot phase was initiated (planned to last until 2024) with the task to transfer those solutions into practice by building a suitable software architecture that allows for real-time subject indexing of economic literature resources with our trained machine learning models and the integration thereof into the other metadata workflows at ZBW. The AutoSE subject indexing service has been in productive operations since 2021 and is continually evolving, with the most current activities pertaining to the introduction of the next generation of language (transformer) models into the model portfolio.

This talk reviews experiences in Finland and beyond on building a series of semantic portal applications in use, based on using a new declarative version of the Sampo-UI framework [1,2] and Sampo Model [3]. The key idea is to re-use a shared semantic web infrastructure [4] and existing user interface (UI) models for fast prototyping and development. In this model, the data service is completely separated from the UI and accessed only with SPARQL querying. It is also shown, how the Sampo model can be applied on top of a cloud of interlinked Knowledge Graphs to create a meta-sampo “SampoSampo” [5] based on other Sampo portals. [1] Esko Ikkala, Eero Hyvönen, Heikki Rantala and Mikko Koho: Sampo-UI: A Full Stack JavaScript Framework for Developing Semantic Portal User Interfaces. Semantic Web – Interoperability, Usability, Applicability, vol. 13, no. 1, 2022. https://doi.org/10.3233/SW-210428 [2] Heikki Rantala, Annastiina Ahola, Esko Ikkala and Eero Hyvönen: How to create easily a data analytic semantic portal on top of a SPARQL endpoint: introducing the configurable Sampo-UI framework. VOILA! 2023 Visualization and Interaction for Ontologies, Linked Data and Knowledge Graphs 2023, CEUR Workshop Proceedings, Vol. 3508, October, 2023. https://ceur-ws.org/Vol-3508/paper3.pdf [3] Eero Hyvönen: Digital Humanities on the Semantic Web: Sampo Model and Portal Series. Semantic Web – Interoperability, Usability, Applicability, vol. 14, no. 4, pp. 729-744, IOS Press, 2023. https://ceur-ws.org/10.3233/SW-223034 [4] Eero Hyvönen: How to Create a National Cross-domain Ontology and Linked Data Infrastructure and Use It on the Semantic Web. Semantic Web - Interoperability, Usability, Applicability, vol. 15, no. 4, 2024. https://doi.org/10.3233/SW-243468 [5] Eero Hyvönen, Annastiina Ahola, Petri Leskinen, Heikki Rantala and Jouni Tuominen: How to Create a Portal for Digital Humanities Research Using a Linked Open Data Cloud of Cultural Heritage Knowledge Graphs: Case SampoSampo. Proceedings: SemDH 2025 Second International Workshop of Semantic Digital Humanities, co-located with ESWC 2025, Portoroz, Slovenia, CEUR Workshop Proceedings, 2025. Forth-coming. https://seco.cs.aalto.fi/publications/2025/hyvonen-et-al-samposampo-semdh-2025.pdf

This paper examines the Brasiliana Guita Library e José Mindlin's (BBM) experience with metadata flow structuring as a component of its digital curation policy. From the integration between the ALEPH/DEDALUS and DSpace systems, via MARC21, Dublin Core Qualified, OAI-PMH, and XSLT, it is discussed how metadata started to act as a technical and cognitive infrastructure for the preservation and access to information. This study is characterized as a descriptive research, based on the documentary analysis of the metadata management processes at BBM Digital The text presents the theoretical foundations of metadata mediation, BBM's internal operational flows and the perspectives of evolution, with emphasis on the FAIR principles and the possibilities of Linked Data as future ways to expand interoperability, the visibility and democratization of access to bibliographic and cultural heritage.

The Food and Agriculture Organization of the United Nations (FAO)’s AGRIS (the International System for Agricultural Science and Technology), has long served as a vital global hub for agricultural and related scientific knowledge. Recently, AGRIS has undergone a series of strategic enhancements aimed at improving data quality, user experience, multilingual access, and interoperability. This presentation will showcase the best practices adopted to strengthen AGRIS as an open-access, inclusive, and high-impact information system. Key improvements include the implementation of a clustering feature to group duplicate bibliographic records, significantly streamlining search results and reducing redundancy. AGRIS has also achieved Digital Public Good (DPG) certification by the Digital Public Goods Alliance, recognizing its commitment to open standards and universal access. To support greater discoverability, a new mechanism has been introduced to better identify records with available full-text resources. The system now offers access to over 15.5 million records in 121 languages, reinforcing AGRIS’s role as a truly multilingual knowledge platform. Additionally, the growing network of new data providers has enriched content diversity and expanded the geographical scope of contributions. The presentation will outline the design and implementation strategies behind these advancements, highlighting the importance of collaboration, metadata harmonization, and inclusive design. It will also explore how AGRIS aligns with the FAIR data principles and supports global efforts toward more equitable and open knowledge dissemination as the leading global repository for agricultural research

AI-powered metadata generation is reshaping knowledge organization but raises critical ethical concerns. This workshop addresses the need for transparency, accountability, and social responsibility in AI-powered metadata practices. Participants will explore real-world risks, identify ethical gaps, and collaboratively develop semantic guidelines for sustainable metadata ecosystems. The proposed framework aims to support responsible AI adoption in linked data and bibliographic ontology contexts.