In this paper we present a project aimed at building a collaborative bibliography on AI and DH topics, promoted by AIUCD’s Special Interest Group DH-AI. The project focuses especially on the subject indexing of bibliography contents, that is, the collaborative tagging of resources. After an overview of the state of the art, we present the DH-AI community’s initial activities and discuss the rationale behind our choices: the use of Zotero as the platform for collecting bibliographic items, and the development of an indexing model based on a controlled vocabulary defined by the community and integrable with existing resources such as the Nuovo Soggettario (NS), the Italian national subject heading system.

The Institute of the Polish Language (Polish Academy of Sciences) holds a collection of lexical resources developed over decades, covering Polish and Latin used in Poland from the Middle Ages to the 21st century. These include monumental dictionaries — among them the Dictionary of Polish Dialects (Karaś et al. 1979–), the Dictionary of Medieval Latin in Poland, the Dictionary of Medieval Polish Personal Names, and the Dictionary of Polish Place Names — as well as card collections totaling approximately 6.5 million slips. This variety presents major challenges to unified storage and retrieval: the resources differ in scope, internal structure, editorial conventions, and metalanguage. In this paper, we present the workflow, data model, and current state of PoliLex, a project aimed at providing integrated access to these resources through harmonized modelling and semantic data publication. The text of most dictionaries was obtained from scans via OCR and post-correction, then progressively structured into TEI-conformant XML. The resulting files serve as the basis for data extraction and mapping to RDF via Ontology-Based Data Access mappings. The data model is built on a project ontology (PoliLex, v. 0.1) extending OntoLex-Lemon (McCrae et al. 2017) and its lexicography module (Khan et al. 2022), with a central distinction between the dictionary as a textual artifact and the language system it describes. The data are accessible through a full-text interface (TEI Publisher) and a SPARQL endpoint exposing lexical data as Linked Open Data, enabling cross-resource retrieval at the level of forms, definitions, and attestations.

This paper presents Zotero RDF Server (ZRS), a modular framework that reimagines the ubiquitous reference manager Zotero as a federated knowledge graph platform tailored to the specific but often shared needs of humanities scholars, especially in settings where large made-to-measure infrastructures are unavailable, slow to deploy, or overly expensive. On a methodological level, the tool presents a potential solution to a notorious problem: a digital reproducible workflow for individual scholars working with scarce resources and with no or little quantitative methods. With ZRS, these scholars not only enhance their work’s replicability by producing interoperable data and queries. The paper argues that scholars also gain digital literacy, taking off from a workflow they use already leading to a Semantic Web representation of their own data.

Zotero is widely used in everyday scholarship for collaborative bibliographic control, note-taking, as well as PDF storage and annotation. Yet its potential as a rich semantic knowledge graph connected to external vocabularies and structured data workflows is still underexplored. ZRS leverages Zotero’s open API to generate RDF triples that retain familiar user practices while adding semantic interoperability, on the fly and with almost no additional effort. The system ingests multiple Zotero libraries, applies configurable RDF/OWL mapping rules to identify entities (e.g., creators, places), and persists them in an embedded Oxigraph triple store that exposes a SPARQL endpoint. Additionally, it performs OCR (Kraken/Transformer) and ingests texts and embeddings in an OpenSearch index for deterministic queries.

Digital Humanities projects often use different tools, standards, and data models. Even when they focus on the same historical period or similar sources, their data usually remains separated. This makes cross-project research difficult.

At the Centre for Medieval Studies at KUL, several projects work on related medieval materials but use different formats, including TEI, wiki and structured databases. To make these resources searchable in one place, we are developing Scrinium, a multisearch application for all research data produced within the Centre.

Scrinium is based on the Typesense search engine and supports full-text search and filtering. The main challenge, however, is not the search engine itself, but the integration of different data models. In order to connect heterogeneous project data, we use Wikibase as a central semantic layer. It serves as a shared core that links entities and concepts across projects.

In this paper, I present the architecture of Scrinium, the strategy for integrating different data models, and the practical problems we encountered. I also discuss how this infrastructure can support the development of a RAG-based system for more precise semantic search.

Scrinium shows how a multisearch environment can be built for diverse DH projects while respecting their individual data structures.

The foundation of an institution in early Europe was supported by legal documents with a rigorous logical and conceptual architecture, endowed with the necessary authority. The proposed research is part of a pilot project that aims to analyze two institutional foundations from the second half of the 16th century during the reign of Stephen Báthory: the Jesuit academy in Vilnius in 1579, validated by royal privilege and confirmed by papal authority, and the act of establishment of the Jesuit College in Cluj (Collegium Academicum Claudiopolitanum, 1581).

The paper analyzes how these acts of institutional legitimation can be structurally represented through TEI encoding and a graph-based modeling approach. The diplomatic architecture of the documents (invocatio, intitulatio, arenga, dispositio, sanctioning clauses) is segmented, special attention being paid to performative verbs (e.g., erigimus, confirmamus) and named entities (persons, institutions, places). The texts provide the basis for the extraction of a knowledge graph representing acts of authority, institutional foundations, and their issuing agencies.

The paper demonstrates how digital formalization allows for both the analysis of a pontifical and a princely foundation within a common Latin-language documentary culture that transcends political and confessional boundaries. The pilot project highlights both the heuristic potential of interoperable coding frameworks and the interpretative decisions required in the process of transforming complex diplomatic documents into relational data structures, and emphasizes the capacity of digital formalisation to model the institutional evolution across the interconnected European political spaces.