Project Reports

The Blue Core project envisions a community-operated and owned BIBFRAME linked data store, where libraries collaborate in the creation and maintenance of linked data for production. As the project continues its work in prototype development, open-source tooling and AI technologies lay the groundwork for future iterations. With open data a pillar of the project, data contributed to Blue Core will be made available under a CC0 public domain dedication and available to all for reuse at no charge.

This report presents results from the project “Systemic Digital Preservation of Archival Records in the Government of the State of Espírito Santo”, which aims to develop a Systemic Digital Preservation (SDP) model for archival records managed in the E-Docs system. The project focuses on constructing a metadata crosswalk among key standards for records management, preservation, and access, with Dublin Core as the central reference. The goal is to support semantic interoperability and enable process automation in digital preservation. Preliminary results include a validated metadata correlation table, with potential applications in interoperability and digital preservation within electronic government environments.

This interdisciplinary project is a collaboration among library metadata librarians, data scientists, digital library technologists, university IT, and the university press. Its goal is to improve the discoverability of academic resources by enhancing metadata through Natural Language Processing (NLP) and embedding-based semantic search, addressing the limitations of traditional keyword-based retrieval. To support this pilot, a library NLP system architecture has been designed, including the development of a vector database to enable semantic search within discovery platforms.

This project presents an innovative approach to analyzing and dating post-1540 Hebrew manuscripts through artificial intelligence. While traditional dating methods rely on manual paleographic analysis requiring extensive expertise, our research leverages MARC metadata from digital manuscript repositories, particularly the KTIV platform containing over 100,000 Hebrew manuscripts. Building on recent advancements in AI applications for historical manuscripts [1, 5] and Hebrew text analysis ]3,4[, we aim to develop systematic methods for manuscript dating and identification of new typological features. Our preliminary analysis of manuscripts from 1600-1700 demonstrates distinct patterns in scribal traditions across different regions and time periods, suggesting the feasibility of AI-powered dating and classification methodologies.

The Respectful Terminology Platform Project (RTPP) is a multi-year collaborative project focused on creating community-focused infrastructure for Indigenous controlled vocabulary in collaboration with the National Indigenous Knowledge and Language Alliance (NIKLA). This article outlines the goals of the project and reports on the current activities of the RTPP.

The Program for Cooperative Cataloging (PCC) Task Group on AI and Machine Learning for Cataloging and Metadata was charged to develop guiding principles, best practices, community of practices, training documentation and plans and international partnership. This report outlines the group's deliverables, methodologies, accomplishments, ongoing initiatives, and the challenges encountered during the initial phase of the project.

The Dublin Core Metadata Initiative has developed a small vocabulary called OpenWEMI. This vocabulary is a minimally constrained abstraction of the concepts of work, expression, manifestation and item (WEMI) that were originally defined in the document "Functional requirements for Bibliographic Records" (FRBR). The WEMI concepts were designed specificially as a conceptual model for library catalog data and are defined with strict constraints on the usage and relationships between the entities. This means that the library-defined model is not extendable to other purposes or communities. And yet, the idea of a model for created resources that can express the levels of abstraction in WEMI has interest outside of the library application. This session will describe how WEMI is being considered in library-adjacent and non-adjacent communities, and how an analysis of current and proposed uses led to the development of a minimally constrained vocabulary for WEMI that can be the basis for a wide variety of metadata creations. OpenWEMI is an RDF vocabulary in the DCMI namespace: https://ns.dublincore.org/openwemi/. Further information is available at https://www.dublincore.org/specifications/openwemi/ and https://journal.code4lib.org/articles/18412.