We present work within a Computational Literary Studies project studying 19th-century poetry in Galician. Computational resources for minority literatures tend to be underdeveloped, and we are helping address this gap for the language. We created automatic metrical scansion tools using symbolic and neural approaches, and encoded an initial TEI corpus (ca.~35,000 lines). We present these and first descriptive findings. An overall goal is exchanging with the community about corpus and computational development in a low-resource setting.

The DoTS suite provides a comprehensive, open-source solution for publishing, searching, and visualizing TEI data. It ensures compliance with FAIR principles while enabling the rapid development of custom web applications tailored to digital humanities and scholarly editing.

The suite includes four main components:

• DoTS Backend: is an implementation of the DTS API in XQuery, powered by the BaseX XML database management system. Compliant with version 1-alpha of the DTS specification, it enables the FAIR publication of TEI sources through the three endpoints defined by the standard: Collection, Navigation, and Document.

• DoTS-vue is a generic and responsive frontend application built with Vue.js, that enables online browsing of text collections served by the DTS API. It integrates the Mirador viewer and offers flexible interface customization options tailored to each corpus.

• DoTS-cli-es is a Python/Flask tool for indexing DoTS collections and documents in Elasticsearch. It exposes a full-text search API that can be consumed by browsing interfaces, enabling efficient and scalable textual queries across corpora.

• ThunderDoTS is a DTS crawler that allows users to query DoTS resources using natural language, paving the way for a more intuitive and accessible search experience within scholarly text collections.

The The Crocodile and Human Mummies of Tebtynis: AI-Assisted Reconstruction and Scholarly Editing of Hellenistic Papyri from the Bancroft Library (Berkeley) project is a joined effort by the Institut für Altertumskunde (IfA), the Cologne Center for eHumanities (CCeH) at the University of Cologne, the Seminar für Alte Geschichte / Forschungsstelle Papyrologie at the University Münster and the Center for the Tebtynis Papyri (CTP) at the University of California, Berkeley. The project is funded by the Nordrhein-Westfälische Akademie der Wissenschaften und der Künste for the upcoming 15 years.

The Bancroft Library in Berkeley houses, with approximately 20,000 objects, the largest collection of papyri from the Hellenistic period. Only around 1,050 of these have been fully or partially catalogued and published over the past 125 years, or are currently being prepared for publication. The far greater part of the collection consists of papyrus fragments waiting to be indexed, edited and (if possible) joined to reconstruct previously unexplored sources.

The nineteenth century saw a dramatic expansion of print, with debates on its effect on linguistic “complexity” often focusing on literary simplification – i.e., shorter sentences, more accessible vocabulary, and convergent stylistic norms – while journalistic and scientific prose may have remained stable or even complexifying. We ask whether comparable patterns are detectable in nineteenth-century Danish newspapers, using a corpus of ~5 million articles (1677–1849), categorized as National news, International news, Advertisements, and Fiction, including serialized narratives. This is a work in progress, where preliminary results of a randomly sampled subset (~200,000 articles) are presented.

Complexity is measured via surface and shallow syntactic features (sentence and word length, lexical diversity) supplemented with features capturing narrative style and syntactic structure. Preliminary analyses indicate domain-structured change: Fiction and International news show the clearest diachronic trends, while aggregate signals are weak. Sentence length declines in Fiction and Advertisements, word length rises across all domains, and lexical diversity increases selectively.

Linear regression predicting publication year performs poorly overall, improving modestly per domain. Our genre-aware pipeline suggests that, rather than general shifts in Danish, international influences may have shaped Fiction, connecting Scandinavian print history to broader European debates on the rise in print, literacy, media, and stylistic change.

The present project takes a transnational perspective on knowledge transfer in the Habsburg monarchy (1848-1918), with a particular focus on translation (Vimr 2025). By digitising historical translation bibliographies and enriching them with Linked Open Data (LOD), we enable, for the first time, the analysis of the structural characteristics of knowledge transfer in a multilingual and multicultural entity such as the Habsburg monarchy, including through data visualisations (Hofeneder 2022). This approach makes it possible to identify asymmetries, hubs, and peripheral zones of knowledge transfer that remain invisible in language-pair or nationally bounded studies.

Rather than focusing on a single language or language pair, we will consider all translations published within the territory of the Habsburg monarchy, in line with a transnational understanding (Judson 2016; Surman 2018). This prevents a nation from being equated with particular languages and, consequently, with fixed territories.

The aggregation of over 100,000 records covering fiction and non-fiction texts reveals structural properties and patterns of knowledge transfer that transcend national frameworks. This method provides a spatial understanding of translation as a dynamic process, revealing movement patterns and relationships that would otherwise be difficult to detect. This graph-based approach significantly enhances our ability to study translation as an interconnected spatial phenomenon.

The ANR-funded (Agence Nationale de la Recherche) project E-cclesia aims to investigate the evolution of discourse surrounding church architecture in the Latin West between the 4th and the 12th centuries. Digital Humanities are at the centre of this multidisciplinary project, which brings together epigraphists, archeologists, philologists, historians and engineers. Indeed, a database linking texts and images related to the research theme is currently being constituted. Moreover, coordinating international teams across institutions and disciplines requires shared governance standards, common ontological frameworks, and structured iterative enrichment processes applied to existing datasets. This poster is conceived as an illustration and reflection on the workflow, Digital Humanities tools, and methods used by the project members.

In recent years, the limited focus on digitally preserving and integrating traditional patterns has resulted in their diminished visibility and use, endangering their continuity within cultural memory and impacting interdisciplinary research across art, history, and digital humanities on a global scale.

The case study explores 2different sections, which have macro to micro vision, the first is traditional pattern traditions from chosen countries. Second is about different areas in these countries that can be redesigned with them.

The methodology for this study will be multi-faceted, combining theoretical research with practical digital modeling and design implementation.

The expected outcome of this study is the development of reinterpreted traditional patterns that can be effectively reused and adapted within digital and contemporary contexts across various application domains.

The novelty of this research lies in its methodological approach, which employs digital twin technologies and three-dimensional modeling of traditional patterns to generate innovative reinterpretations.

The result is to create environments that maintain a connection to cultural identity while embracing contemporary design and technological innovation, which helps to preserve cultural heritage.

We present a multilingual dataset of fiction in 7 languages, enriched with manual annotation about characters and events in the stories. The dataset includes full stories of diverse lengths, ranging from 500 to 17k words, with a total of around 100k words per language. The corpus was primarily compiled from three online fiction platforms: Archive of Our Own (AO3), Postype, and Wattpad. We collected texts in seven languages (Bahasa Indonesia, Chinese, Dutch, English, Italian, Korean, and Spanish). To maintain a balanced and diverse corpus across languages, we sampled works along multiple dimensions, including fandom, worldbuilding settings, narrative perspective, dialogue/narration ratio, genre, and popularity (kudos/hits ratio). All data, guidelines, and annotations will be openly released.

This contribution presents the Open Encyclopedia System (OES), an infrastructure designed for innovative academic publishing in the humanities and social sciences. The open-source software, developed in close collaboration with researchers since 2020, enables them to autonomously create and maintain open-access publications, such as online encyclopaedias, lexicons, living handbooks and curated knowledge portals containing multilingual and multimodal content, in accordance with scholarly publishing standards.

Thus, OES broadens the open access publication landscape beyond online journals and monographs to include a new type of open publication. Through case studies from the humanities, history, and the social sciences, we introduce the key features of the OES software and demonstrate its potential for author-driven online publications, particularly those involving content subject to constant review, rewriting, and supplementation. OES is built on the open-source content management system WordPress, enhancing its functionality to support lemma-based academic publications.

Given the specificity of innovation within Social Sciences and Humanities (SSH), this poster presents the OPERAS Innovation Lab, contributing to discussions on how scholarly infrastructures can support innovative DH practices. The Lab’s operational model was developed through iterative observation and analysis of real use cases in SSH scholarly communication. This evidence-based approach shaped its dual structure: the Observatory, mapping innovation dynamics, and the Accelerator, supporting the development of promising ideas. Drawing on Horizon Europe–funded GRAPHIA project, the poster will present Lab's methods to support pilots integrating AI, knowledge graphs, and large language models in SSH workflows. By presenting its operational model, methodology, and practical implementation of its support, the Innovation Lab positions itself as a collaborative initiative devoted to strengthening innovation capacity in the SSH community.

The Czech Republic has developed a national ecosystem of large research infrastructures under the Ministry of Education, Youth and Sports, formalised in the Roadmap of Large Research Infrastructures of the Czech Republic. Four infrastructures serve the humanities: the Czech Archaeological Information System, Czech Literary Bibliography, Czech National Corpus, and LINDAT/CLARIAH-CZ. Their disciplinary specificities pose challenges for interoperability, sustainability, and community adoption.

All participate in Open Science II (OS II; 2025–2028), establishing the Czech national node within the European Open Science Cloud (EOSC) and promoting FAIR principles tailored to humanities-specific needs. The Humanities and Arts cluster is coordinated by a mini-consortium representing these RIs.

Activities focus on developing new repositories, enhancing existing ones, and interconnecting with national services such as the National Metadata Directory and National Repository Platform. This approach combines shared national services, discipline-specific layers, and community-driven governance, balancing coordination with autonomy.

The Czech model provides a scalable framework for interoperable, sustainable, and collaboratively governed humanities infrastructures aligned with EOSC and FAIR principles.

Very often, if only implicitly, the critical apparatus records grammatical considerations. Their digital encodings usually confine themselves to classifying variants by their grammatical differences from the lemma and even omit some implicit data. The poster will first survey the plethora of grammatical information incorporated into printed critical editions. Then, it will present early work on encoding that information using the TEI morphosyntactic and (proposed) syntactic annotations, satisfying formulated desiderata. These will allow for semi-automatic classification of variants and disambiguate their representation. The poster will also try to develop tools for the interaction between linguistic annotations and the critical apparatus. Finally, it will highlight further opportunities for employing the information from the encoding.

Place in GIS data models and ontologies is mostly handled via a vague concept of place which hinders data ingegration and proper handling of place as object of research. As a solution we propose an ontology design pattern based on the CIDOC CRM extension SDHSS (Semantic Data for Humanities and Social Sciences) (Beretta 2024) in order to model place according to the definition of place as ``meaningful site that combines location, locale, and sense of place'' (Cresswell 2009). In order to test this pattern for different cases studies from different disciplines, a ResearchSpace knowledge contextualization environment is currently developed at the Service Center for Digital Humanities (SCDH) in Münster. The experimental platform is used in the Working Group Data Modeling of the Digital Humanities Circle (AK DH) at the University of Münster focusing on case studies and examples of historical town types and the social ontology of early modern places. After an evaluation phase, the ResearchSpace environment will be opened as experimental and explorative platform beyond academic institutions in Münster. Our long-term goal is to provide a semantic platform to enable researchers to collect, share, explore, curate, and reuse place-based information about historical places throughout Early Modern Europe and beyond.

Positioned in the area of corpus stylistics and translation, this corpus-based multifactorial study aims to explore how Lithuanian translators handled repeatedly used reporting verbs signalling direct speech in English-original literary novels. Using the English-original novel "Brave New World" by Aldous Huxley, extracted from InterCorp corpus v.15 (Rosen et al. 2022) and its Lithuanian translation by Gražvydas Kirvaitis (2018), we conducted a two-stage corpus-based study focusing on, first, identification of translation patterns found in the novel, and, second, on exploration of factors (orthographic, syntactic, semantic and pragmatic) that impacted preservation or avoidance of the original’s repetitions in the Lithuanian translation. We manually annotated 315 bilingual concordances with 3rd person past tense reporting verbs, extracted using CQL queries, and their Lithuanian counterparts with respect to five potential predictors of repetition shift. The results of mixed-effects logistic regression conducted in the R environment (R Core Team 2025) show that the translator tends to preserve source-text repetitions when the narrative is left structurally and pragmatically unchanged. However, when the translator intervenes more substantially, by shifting stance, reworking syntax, or when source-text reporting verbs were accompanied by adverbial modifiers, there is a strong tendency to break source-text repetition patterns and increase lexical variety in translation.

Researchers repeatedly encounter questions while working with research data that they themselves describe as “ethically and legally challenging.” The expert group ELSA4Memory from the German National Research Data Infrastructure (NFDI) consortium dedicated to history and historically oriented fields gives answers by picking up "real live" legal and ethical questions from digital (research) projects. The poster presents their work and interacts with participants to start the discussion on how ethical and legal guidelines can best reach a majority of academia.

Data practices often rely on assumptions of objectivity and technologies that obscure power structures. Those assumptions are critically challenged by feminist frameworks. This poster asks how feminist data practices can inform decisions about digitization and platform choice in ethnographic research.

The project is based on a data sprint conducted in a graduate seminar, in which ethnographic data were collected on the planned construction of hyperscalers in the former coal mining city of Grevenbroich, Germany. The collaboratively produced issue map raised the question of how the data could be sustainably preserved and further analyzed without reproducing objectifying or extractive data logics.

Drawing on feminist data approaches, care-based perspectives, and a critical engagement with FAIR and CARE principles, the poster proposes a “care check” for digital platforms used in ethnographic research. It highlights two platforms in particular: nodegoat, a web-based research environment, and the Platform for Experimental Collaborative Ethnography (PECE), an open-source project for collaborative analysis and publication.

The poster argues that platform choice must be treated as an epistemic and political decision and develops criteria that can be applied more broadly to digital research infrastructures in order to foster fair and care-oriented data practices.

The documentary series “Risposte” (Archivio di Stato di Venezia) forms an almost continuous collection of advisory opinions drafted by the Cinque Savi alla Mercanzia for the Venetian Senate (1540-1733). The paper presents a pilot study that treats the “Risposte” as structured units of policy reasoning that can be modelled and analysed at scale. Focusing on register 142 (1607-1610), for which complete digitizations and a subset of manual diplomatic transcriptions are available, we define the single “Risposta” as the primary decision unit and design a two-layer data model: a minimal documentary core plus an analytical layer with controlled vocabularies (policy domain, decision orientation, triggers, actors, geography). Institutional variants are encoded as explicit relations rather than treated as exceptions. To support scalability, we evaluate an ATR/HTR workflow and we prototype lightweight NLP procedures (deontic formula extraction, monetary expressions, similarity measures) for semi-automatic population and quality control. The pilot operationalises two directions of scalability: a stable decision-unit model and an interoperable transcription pipeline to populate it, enabling future large-scale analysis of institutional coherence, policy drift, and risk management in early modern Venice.

Projekt realizowany w ramach PF Digital Humanities Lab (ID UJ) ma na celu stworzenie cyfrowego korpusu oraz naukowej edycji staropolskich przekładów Księgi Psalmów. Obejmie on wybrane psalmy ze średniowiecznych psałterzy rękopiśmiennych (m.in. Psałterza floriańskiego i Psałterza puławskiego) oraz wersji drukowanych: katolickich i protestanckich wydanych do 1632 roku. Dobór psalmów uwzględnia ich różnorodność gatunkową i ich znaczenie dla recepcji literackiej.

Korpus powstanie jako edycja równoległa, umożliwiająca zestawianie różnych przekładów, wariantów tekstu oraz porównanie z Wulgatą i przekładem współczesnym (Biblia Tysiąclecia). Użytkownikom zostanie udostępniony m.in. widok transkrypcji i faksymile, aparat krytyczny, synchronizacja segmentów tekstu, interaktywny interfejs oraz anotacje semantyczne i komentarze filologiczne. Projekt ma charakter dydaktyczny i będzie wykorzystywany w nauczaniu historii języka, gramatyki historycznej, literatury, tekstologii i przekładoznawstwa.

Teksty zostaną zakodowane w standardzie TEI, co umożliwi wielopoziomowe oznaczenie struktury i znaczeń oraz dalsze przetwarzanie danych. Kluczowym wyzwaniem jest modelowanie danych integrujące liczne warstwy anotacji, zachowanie relacji między różnymi systemami numeracji psalmów oraz opracowanie dostosowanej do dawnej polszczyzny anotacji morfosyntaktycznej. Prezentacja i wizualizacja materiału zostaną zrealizowane z wykorzystaniem oprogramowania TEI Publisher (Jinks). Projekt proponuje nowy model dydaktycznej edycji cyfrowej, łączącej funkcje badawcze i edukacyjne. Posiada potencjał adaptacyjny w innych projektach humanistyki cyfrowej, wyznaczając kierunek rozwoju edycji tekstów dawnych w środowisku interoperacyjnych standardów.

Digital text-based research in the Digital Humanities increasingly relies on a diverse ecosystem of specialised tools, repositories, and computational environments. Despite this richness, interaction between these components often remains fragmented. Transitions between analysis, interface, data management, and publication are particularly challenging: repositories typically function as endpoints, user interfaces are developed on a project-specific basis, and analytical workflows are often tied to local or ephemeral environments. This fragmentation not only complicates the reproducibility of research but also introduces inefficiencies and barriers to collaborative work, long-term data management, and iterative publication practices.

The poster will visualise new developments for the TextGrid repository, highlighting the integration of repository services, notebook-based workflows, and the Jupyter4NFDI infrastructure. Rather than presenting Jupyter merely as a notebook environment, the contribution conceptualises it as an infrastructural binding layer connecting analysis, interface, repository access, and process-oriented publication practices. By illustrating key stages of development, system components, and exemplary workflows—including the publication pipeline and on-demand text processing—the poster demonstrates how Jupyter functions as a sustainable interface infrastructure that structurally supports reuse, further processing, and iterative publication of text-based research data within a coherent, scalable architecture.

This poster presents the first practical results of the FinFAIR project funded through the ECHOES Cascading Grants programme. As one of the first selected initiatives delivering structured Heritage Digital Twins to the emerging European Collaborative Cloud for Cultural Heritage, we demonstrate how small-scale, multilingual heritage ecosystems can integrate into European digital infrastructures without losing institutional autonomy or cultural specificity.

The Finno-Ugric Data Sharing Space is a federated public–private semantic infrastructure built with Wikibase and aligned with the European Interoperability Framework. Rather than centralising collections, it connects library, museum, archival, and field research workflows into a shared interoperability layer. Heterogeneous metadata (MARC/DCTERMS, CIDOC CRM, archival finding aids) is reconciled through identifier alignment (ISNI, VIAF, GeoNames), multilingual modelling for agglutinative languages, and component-based digital twin structures linking tangible and intangible heritage.

The DSS serves as a staging environment for ECCCH while increasing community visibility through selective synchronisation with Wikimedia Commons, Wikidata, and Wikipedia. A Livonian music case study shows how fragmented archival recordings were transformed into interoperable digital twins capable of modelling layered rights regimes, including author, performer, moral, personal data protection, and CARE considerations.

Beyond technical integration, the project incorporates cross-border reconciliation and community validation, demonstrating how governance-aware modelling enables epistemic restitution alongside interoperability.

The poster introduces a modular framework for open, sustainable, and community-curated tool registries built upon the well-established and open infrastructures of Wikimedia, GitLab, and Zenodo for creating, maintaining, sharing, curating, and archiving linked open data.

The application of standardized labels to categorize content (e.g. by PEGI) can be considered established, sometimes even legally required, practice for marketing digital products of art (literature, films, games, music etc.). In that sense, today‘s sources of artistic practice (as potentially: tomorrow‘s objects of historical interest) often already bear explicit information regarding their moral/ethical nature, which are not necessarily fully included into respective metadata. Further, not only artistic but also scientific publication contexts increasingly react to the same idea of content-labelling (e.g. with symbols in repositories or by phrasing in-text „content notes“/"trigger warnings", especially for students).

Despite psychological details of the process being scientifically contested, a broad consensus about the overall positive effects of carefully administered content notes may be assumed. So research data and more specifically: a metadata scheme for describing research data should be prepared for the need to express content warnings, which are considered too important to just be implied.

This submission discusses a few techniques on how to balance data providers' responsibility to protect their users' interests, against the perception of scholarly paternalism, sometimes associated with "warnings" in academic contexts. It does so with the aim of improving existing representational strategies for content notes in metadata.

The poster presents a development project on a theatre studies source that, despite its central value for literary studies and cultural history, has not yet been digitised. The ‘Deutscher Bühnen=Spielplan’, a stage programme almanac, was published between 1896 and 1944, recording all the programmes of German-language theatres in the form of calendar lists and thus documenting the reality of modern theatre performances.

The focus is on processing the volumes printed in Gothic script up to 1910. Their complex layout can be captured using the advanced layout analysis functions of Transkribus (combination of field and table models) in order to use line and text recognition models for automatic transcription of the table contents in the next step. The long-term goal is to create a web application that will fully reflect the information density of the ‘Bühnen=Spielplan’, which also contains details such as translation versions and guest performances. At the same time the project serves as an example of the application of AI for the development of similar sources that are relevant to cultural history but challenging in terms of layout.

Oral history and community memory are core resources for historical inquiry, yet spatial and material aspects of remembered scenes can be difficult to externalize and compare when they circulate primarily through verbal exchange. This poster proposes an Iterative AI-Assisted Framework for Visual Reconstruction and Memory Negotiation that uses generative AI not to verify memory or produce definitive reconstructions but to create provisional visual 'probes' that support discussion, revision, and comparison. Grounded in oral history and memory studies and informed by digital humanities critiques of visual authority, the workflow proceeds in five stages: (1) narrative elicitation; (2) generative visual prototyping; (3) participant-led iterative revision (human-in-the-loop); (4) multi-narrator comparison and negotiation; and (5) a negotiated reconstruction archive that preserves final images, intermediate iterations, and records of agreement, uncertainty, and disagreement. A vignette illustrates the method through two elders recalling a no-longer-extant building. The poster highlights ethical and methodological constraints, especially the tendency of compelling images to anchor interpretation, and treats outputs as interpretive artifacts rather than evidence. By archiving negotiation traces alongside visual outputs, the approach supports reflexive public-history engagement and responsible documentation in contexts of visual archive scarcity.

Tematem posteru jest prezentacja wyników najnowszych badań nad fleksją najstarszej polszczyzny, które są prowadzone w Pracowni Języka Staropolskiego IJP PAN. W latach 2018-2024 zrealizowano projekt Baza leksykalna średniowiecznej polszczyzny. Fleksja. W efekcie powstała internetowa baza danych składająca się z trzech aplikacji: Słownika, Korpusu i Biblioteki.

Baza gromadzi informacje na temat morfologii wyrazów poświadczonych w zabytkach językowych powstałych do około 1500 r., głównie w oparciu o materiał Słownika staropolskiego wraz z suplementami. Aplikacja Słownik składa się z haseł, które zawierają wszystkie odnotowane formy fleksyjne danego leksemu. Dzięki tej formie gromadzenia, przechowywania i udostępniania danych możliwe jest wyszukiwanie przykładów określonych form fleksyjnych oraz ich analiza. Formy są ilustrowane cytatami najstarszych odnotowanych poświadczeń. Ponadto hasła zawierają informacje semantyczne i frazeologiczne, zwykle czerpane ze Słownika staropolskiego. Dostępne są również skany odnośnych stron Słownika. W Blśp zamieszczono również szczegółowe informacje o źródłach wszystkich cytatów. Odnotowane zostały także zmiany, których dokonano w obrębie haseł w stosunku do Sstp i jego suplementów.

W Bibliotece Blśp zgromadzone zostały wybrane teksty ciągłe udostępnione w transliteracji i transkrypcji, co umożliwia ich pełnotekstowe wyświetlanie.

Blśp zawiera także zalążek korpusu polszczyzny do około 1500 r.

Podczas wystąpienia zostaną zaprezentowane podstawowe funkcje Blśp, możliwości jej zastosowania w badaniach historycznojęzykowych oraz perspektywy rozwoju projektu.

Archival research in the digital age is frequently characterized by "tool fatigue", where scholars must navigate a disconnected archipelago of software solutions. A historian might store scans in a commercial cloud, manage citations in a desktop manager, perform OCR in a proprietary "black box", and attempt visualizations in yet another isolated platform. This disjointed approach leads to data silos, loss of context, and significant friction when moving between individual stages of research. Furthermore, the reliance on proprietary ecosystems often compromises the long-term sustainability and reproducibility of research data, which represent the core values of the FAIR principles and DH communities.

The proposed poster addresses these systemic issues by demonstrating a unified technical pipeline that is modular, open-source, and accessible. Utilizing OwnCloud, Zotero, Arkindex, Nodegoat, and Logseq it moves beyond the demonstration of single tools to focus on the connectivity between them. The core idea of the proposed pipeline is interoperability: ensuring that data flows seamlessly from on-site scanning to final visualization via standard protocols (WebDAV, APIs) and open formats (Markdown, JSON, XML). The pipeline advocates for a shift from ad-hoc tool adoption to a structured, engineered approach to humanities data.

This poster presents an interoperability framework developed within the LOD4HSS initiative to connect OntoME, an online ontology and application profile documentation platform, with WissKI, a Linked Open Data production environment based on semantic paths. The framework bridges the gap between conceptual modelling and operational implementation in digital humanities infrastructures.

OntoME supports the collaborative documentation and versioning of ontologies and namespaces. Within LOD4HSS, documented conceptual models can be formalised in OWL and made accessible via APIs, allowing WissKI to retrieve classes and properties and activate its reasoning mechanisms. This ensures that implemented data models are explicitly grounded in documented semantic decisions.

A second development concerns the conceptual documentation of semantic paths, a core feature of WissKI. By enabling the description, justification, and versioning of paths in OntoME, these modelling structures become reusable, citable objects that can be exported in the WissKI pathbuilder format and implemented across projects.

The proposed workflow enhances interoperability, reuse, and sustainability, and is replicable in other path-based systems, contributing to more transparent and sustainable semantic infrastructures for the humanities.

The contribution introduces Recollectio - a framework for creating digital collection catalogs. While still in its implementation phase, it stands out by the approach adopted with respect to the data, considered not only as a layer underlying and informing the catalog proper, but as a publicable object - a research dataset - equal in significance to the catalog itself.

Building on XML as the inner data format Recollectio allows to encode both structured data and annotated texts in a unified manner, which can easily be shared and reused as XML is a non-proprietary open format. Furthermore, Recollectio encourages and facilitates adopting Linked Open Data principles.

When it comes to the online publication of the catalog, Recollectio adheres to the mininal computing approach by generating user-friendly internet sites with low infrastructure requirements. This way, the online publication barrier gets lowered, making it easier both for organizations and individual scholars to publish catalogs online and make them widely available to the scientific community and general public.

This poster presents a project reconstructing the information network of scientific publishing in Czechoslovakia (1910–1938). It maps the development of science in a newly established state facing tensions between national independence and international trends. Integrating Digital Humanities and Information Science, the project views science as a complex process of knowledge circulation, overcoming traditional biographical narratives through macroscopic modeling.

Methodologically, the research relies on Social Network Analysis (SNA) and distant reading of a large corpus of digitized scholarly periodicals. The application of infometrics and specifically SNA allows for an examination of scholarly communication as a macroscopic model of thematic, institutional, and personal relationships. The research addresses three methodological challenges in processing historical texts:

- Automated document segmentation of OCR data lacking structural metadata.

- Named Entity Recognition (NER) for identifying authors and institutions within graph databases (Neo4j).

- Comparing traditional (LDA) and neural (BERTopic, LLMs) topic modeling for interpreting historical scientific terminology.

The poster presents a pilot data sample through an interactive demonstration. Visitors can explore the modeled network, visualize topic transformations, and discuss the limitations of automated OCR processing in historical research. This study provides critical insights into the interplay of culture, politics, and scientific discourse.

This poster presents a reference network for early modern natural philosophy (c. 1600–1800). Unlike modern citation networks, early modern texts lack standardized references, yet in-text mentions reveal how scholars positioned themselves during a period of scientific transformation. These references, extracted from over 400 Latin works, allow us to map intellectual engagement and construct a historical reference network.

Processing unstructured, noisy data required a custom multilingual transformer-based Named-Entity-Recognition (NER) model, as no existing tool supported early modern Latin. The analysis identified 50,000 mentions and 80,000 ties, highlighting the prominence of ancient and Arabic scholars and a shift toward experimental and Newtonian natural philosophy.

The historical approach to citation network analysis reveals a field shaped by both continuity and change: ancient and scholastic authorities remained central, while figures like Descartes, Newton, Sennert, Gassendi, and Mariotte emerged as influential.

The poster features a network graph, source and tool criticism, and a critical reflection on the research’s insights and limitations. By linking European scholars and extending beyond their boundaries through data-driven, machine learning methods, this study aligns with the EADH theme, offering a critical examination of intellectual exchange and the evolution of natural philosophy.

I present a toolkit for tracing information flows in nineteenth-century newspapers by identifying sources, intermediaries, and authors across multiple digitised collections. Motivated by rumours such as the “Great Moon Hoax,” the methods generalise to broader questions about how news and scientific claims circulated across national and linguistic borders. The toolkit combines three elements: (1) a cross-collection gazetteer of historical newspaper titles; (2) detection of explicit sourcing and implicit text reuse; and (3) an inspectable network view of inferred flows. I build a title authority list by harmonising catalogues from impresso, Chronicling America, and British Library/British Newspaper Archive sources, normalising title variants and resolving duplicates via fuzzy matching and shared metadata. The gazetteer anchors short, ambiguous references (e.g., “according to the Herald”) to concrete newspapers even when titles appear in variant or OCR-degraded forms. For explicit sourcing and signatures, I combine cue patterns (including multilingual equivalents), lightweight named-entity recognition, and positional/formatting heuristics. For implicit transmission, I compute n-gram similarity over candidate article pairs and use publication dates to propose directionality. Relations are stored in a graph model and explored through notebook-based filtering and path inspection, supporting historical questions about media geography, multilingual circulation, and the politics of reprinting.

This poster presents the Database of Early Modern Czech Book Illustration (e-ilustrace.cz) as a digital humanities infrastructure reshaping approaches to Czech printed illustrations of the early modern period. Bringing together more than 6,600 woodcut illustrations from Czech prints up to the mid-sixteenth century, the project moves beyond digitization toward computationally structured visual scholarship and large-scale image analysis.

Three features make the resource particularly relevant for current DH research. First, its structured iconographic intelligence: records integrate detailed bibliographic description with ICONCLASS classification, enabling complex thematic and comparative queries across dispersed corpora. Second, its IIIF-based interoperability allows high-resolution viewing, side-by-side comparison, and integration with external repositories, situating Czech material within a transregional visual network. Third, it provides tools for analytical reuse tracking, making it possible to trace individual woodblocks across multiple editions and thus reconstruct workshop practices, image circulation, and serial visuality in early modern print culture.

Rather than functioning as a static repository, e-ilustrace operates as a research laboratory—illustrating how standardized metadata, visual technologies, and interoperable infrastructures can support new methodological perspectives in book history and early modern visual studies.

Aplikacja „Korpusuj”, zaprezentowana pierwotnie podczas LXXXII Zjazdu Polskiego Towarzystwa Językoznawczego, jest narzędziem przeznaczonym do samodzielnego tworzenia, tagowania i przeszukiwania korpusów przy użyciu modeli językowych wchodzących w skład bibliotek do przetwarzania języka naturalnego (SpaCy, Stanza). Przedmiotem posteru jest prezentacja najnowszego etapu rozwoju oprogramowania, ukierunkowanego na rozszerzenie jego możliwości o zautomatyzowaną analizę kolokacji oraz semantyczną identyfikację struktur tekstowych.

Program został rozbudowany o moduł kolokatora. Umożliwia on ekstrakcję i analizę łączliwości wyrazów w badanym materiale korpusowym. Funkcjonalność ta umożliwia identyfikowanie istotnych statystycznie współwystąpień jednostek leksykalnych. Ponadto składnia języka zapytań została wzbogacona o nowe operatory umożliwiające zawężenie wyników wyszukiwania w zależności od współwystępowania tokenów w określonym oknie kontekstowym.

W warstwie przetwarzania języka naturalnego (NLP) zaimplementowano mechanizmy rozpoznawania koreferencji. Język zapytań korpusowych wzbogacono o funkcję identyfikacji i filtrowania jednostek należących do konkretnych klastrów koreferencyjnych, co ułatwia badanie spójności tekstu i śledzenie referentów.

Modyfikacje objęły również graficzny interfejs użytkownika, który dostosowano do pracy z nowymi warstwami analitycznymi. Widok pełnotekstowy zyskał funkcję podświetlania jednostek nazwanych (NER) oraz klastrów koreferencyjnych. Wprowadzono także możliwość generowania wizualizacji w postaci grafów zależności składniowych dla wyszukiwanych wypowiedzeń. Ponadto dodany został graficzny konstruktor zapytań, który stanowi alternatywę dla ręcznego wpisywania kwerend.

The G. Calinescu Institute of Literary History and Theory in Bucharest (Romanian Academy) established, in 2026, a digital philology laboratory, integrated into the Critical Editions Department. The first initiative of this working group (made up of George Neagoe and Mihăiță Stroe) was to create, in collaboration with the computer scientist Adrian Cosmin Popescu, a digital working framework. The purpose of this tool is to build a corpus of critical digital editions of the classic authors of Romanian literature: Mihai Eminescu, I.L. Caragiale, Tudor Arghezi, Mircea Eliade, Camil Petrescu and others. Thus, we aim to offer an image that goes beyond the authorial centrality in scholarly editing. We are also interested in showing, by comparing multiple windows, how a text has changed from one edition to another. In addition to this, we try to highlight, through genetic criticism, how various ideologies (communist propaganda, censorship, rewriting) participated in the reconfiguration of drafts and manuscripts. At this moment, our study is concerned with Camil Petrescu’s novel "Un om între oameni" ["A Man Among Men"], about Nicolae Bălcescu, an important figure of the Revolution of 1848. It was written between 1950-1957, in the Stalinist period of Romania.

The paper explores the possible operationalization of the concept of “Otherness”, by testing how effectively Large Language Models can identify its representation in literary texts. Using Joseph Conrad’s Heart of Darkness as an initial case study, we compared the performance of GPT-5 and LLaMA 3 against a human-curated benchmark. Initial results suggest that prompt engineering is key to helping models distinguish between neutral passages and those representing the "Other." This study constitutes the groundwork for a wider project, which aims to scale this methodology across a vast corpus.

Referat poświęcony jest projektowi RE-CREATIO - dziedzinowemu repozytorium cyfrowemu gromadzącemu niezrealizowane projekty architektoniczne. Punktem wyjścia jest napięcie między koniecznością ochrony zbiorów specjalnych a misją ich upowszechniania. Rękopisy, rysunki i projekty architektoniczne, jako nośniki pamięci kulturowej, są szczególnie podatne na zniszczenie, dlatego dostęp do nich bywa ograniczony. W efekcie wiele cennych obiektów pozostaje poza społeczną świadomością.

Odpowiedzią na ten problem jest digitalizacja i tworzenie repozytoriów umożliwiających bezpieczne, szerokie udostępnianie zasobów. RE-CREATIO koncentruje się na projektach, które nigdy nie zostały zrealizowane –- zarówno utopijnych wizjach teoretycznych, jak i koncepcjach przeznaczonych do wykonania, lecz zaniechanych z przyczyn politycznych, finansowych czy historycznych. Wśród nich znajdują się m.in. plansze Villi Laurentiny według koncepcji Stanisława Kostki Potockiego, cenotaf Izaaka Newtona autorstwa Étienne’a-Louisa Boullée czy projekt Kościoła Najwyższej Opatrzności Bożej Jakuba Kubickiego.

Repozytorium nie ogranicza się do prezentacji reprodukcji. Każdy projekt opracowywany jest jako uporządkowany zbiór obiektów opatrzonych szczegółowymi metadanymi, informacjami o kontekście historycznym i biogramami twórców. Struktura zgodna z międzynarodowymi standardami zapewnia interoperacyjność i badawcze wykorzystanie danych.

Wystąpienie przedstawi genezę i założenia RE-CREATIO, a także wyzwania związane z identyfikacją rozproszonych materiałów i ich krytycznym opracowaniem. Celem referatu jest ukazanie, że cyfrowe repozytoria mogą skutecznie łączyć ochronę dziedzictwa z jego popularyzacją, stając się przestrzenią interdyscyplinarnego dialogu i narzędziem budowania nowych narracji o architekturze, która nigdy nie zaistniała w przestrzeni realnej.