@inproceedings{gradl2026_117723,
  title = {Future Applications : The Text+Registry as Reusable Technical Component Beyond Resource Cataloguing},
  author = {Gradl, Tobias and Fruth, Leon and Geißler, Nils and Hensen, Kilian Erasmus and Schulz, Daniela},
  year = {2026},
  pages = {166-174},
  publisher = {Otto-Friedrich-Universität},
  address = {Bamberg},
  urn = {urn:nbn:de:bvb:473-irb-117723x},
  url = {https://fis.uni-bamberg.de/handle/uniba/117723},
  language = {eng},
  abstract = {Digital humanities research software and infrastructures are typically designed around highly specific scholarly requirements. While effective for particular research practices, they often lack transferability across domains. The Text+ Registry, developed within the Text+ consortium of the German National Research Data Infrastructure (NFDI), proposes an alternative approach. Rather than serving solely as a catalogue for editions, lexical resources, and collections, it is conceived as a reusable infrastructural component adaptable to diverse scholarly and institutional contexts.
This paper examines three architectural principles underlying the Registry: metamodelling, model-driven design, and information layering.
Furthermore, two application scenarios demonstrate broader reuse: a transnational meta-archive of German archival materials on modern India (MIDA) and a planned NFDI-wide helpdesk catalogue for research data management services.
Technically, the Registry harvests distributed data via standard protocols while aligning heterogeneous schemas through configurable mappings.
The Text+ Registry thus models an infrastructural philosophy that supports pluralism: different conceptual frameworks can coexist and interoperate without enforced standardization. This approach reconciles disciplinary specificity, interoperability, automation, and curation, offering a sustainable model for digital humanities infrastructures and potentially for research infrastructures beyond the humanities.},
  keywords = {Text+, Text+ Registry, Resource Catalog, Research Infrastructure, NFDI, MIDA}
}

@inproceedings{gradl2026_117769,
  title = {Future Applications : The Text+Registry as Reusable Technical Component Beyond Resource Cataloguing},
  author = {Gradl, Tobias and Fruth, Leon and Geißler, Nils and Hensen, Kilian Erasmus and Schulz, Daniela},
  year = {2026},
  booktitle = {European Association for Digital Humanities 2026 : "Linking Europe: Digital Humanities Without Borders"},
  pages = {166-174},
  publisher = {Zenodo},
  isbn = {978-83-977695-1-9},
  doi = {10.5281/zenodo.23363983},
  url = {https://fis.uni-bamberg.de/handle/uniba/117769},
  language = {eng},
  abstract = {Digital humanities research software and infrastructures are typically designed around highly specific scholarly requirements. While effective for particular research practices, they often lack transferability across domains. The Text+ Registry, developed within the Text+ consortium of the German National Research Data Infrastructure (NFDI), proposes an alternative approach. Rather than serving solely as a catalogue for editions, lexical resources, and collections, it is conceived as a reusable infrastructural component adaptable to diverse scholarly and institutional contexts.
This paper examines three architectural principles underlying the Registry: metamodelling, model-driven design, and information layering.
Furthermore, two application scenarios demonstrate broader reuse: a transnational meta-archive of German archival materials on modern India (MIDA) and a planned NFDI-wide helpdesk catalogue for research data management services.
Technically, the Registry harvests distributed data via standard protocols while aligning heterogeneous schemas through configurable mappings.
The Text+ Registry thus models an infrastructural philosophy that supports pluralism: different conceptual frameworks can coexist and interoperate without enforced standardization. This approach reconciles disciplinary specificity, interoperability, automation, and curation, offering a sustainable model for digital humanities infrastructures and potentially for research infrastructures beyond the humanities.},
  keywords = {Text+, Text+ Registry, Resource Catalog, Research Infrastructure, NFDI, MIDA}
}

@inproceedings{lachner2026_117762,
  title = {Support for Reuse in Data Journals : Current State and Future Perspectives for the Humanities},
  author = {Lachner, Sonja and Fruth, Leon and Henrich, Andreas},
  year = {2026},
  pages = {1-5},
  publisher = {Otto-Friedrich-Universität},
  address = {Bamberg},
  urn = {urn:nbn:de:bvb:473-irb-117762x},
  url = {https://fis.uni-bamberg.de/handle/uniba/117762},
  language = {eng},
  abstract = {Research data reuse is an objective of Open Science, and data journals have emerged as an important publication format for supporting reuse of research data. However, it remains unclear how this objective is operationalized in humanities-oriented data journals and whether reuse-related information is made accessible through their search interfaces. This paper investigates the support for research data reuse in humanities-oriented data journals. We synthesize prior work on research data reuse and identify diferent reuse purposes and forms. In addition, we analyze how ten humanities-oriented data journals implement support for reuse in their documentation structures and search interfaces. Our analyses show that reuse is frequently recognized as an objective in humanities-oriented data journals, but reuse-related information is not directly represented in the search interfaces. Building on these fndings, we derive implications that address the representation of reuse-related information and the design of search systems.},
  keywords = {Research Data Reuse, Humanities Research Data, Data Journals}
}

@inproceedings{lachner2026_117761,
  title = {Support for Reuse in Data Journals : Current State and Future Perspectives for the Humanities},
  author = {Lachner, Sonja and Fruth, Leon and Henrich, Andreas},
  year = {2026},
  booktitle = {JCDL '26: Proceedings of the 26th ACM/IEEE Joint Conference on Digital Libraries},
  pages = {1-5},
  publisher = {ACM},
  address = {New York, NY; United States},
  isbn = {979-8-4007-2597-5},
  doi = {10.1145/3805696.3846475},
  url = {https://fis.uni-bamberg.de/handle/uniba/117761},
  language = {eng},
  abstract = {Research data reuse is an objective of Open Science, and data journals have emerged as an important publication format for supporting reuse of research data. However, it remains unclear how this objective is operationalized in humanities-oriented data journals and whether reuse-related information is made accessible through their search interfaces. This paper investigates the support for research data reuse in humanities-oriented data journals. We synthesize prior work on research data reuse and identify diferent reuse purposes and forms. In addition, we analyze how ten humanities-oriented data journals implement support for reuse in their documentation structures and search interfaces. Our analyses show that reuse is frequently recognized as an objective in humanities-oriented data journals, but reuse-related information is not directly represented in the search interfaces. Building on these fndings, we derive implications that address the representation of reuse-related information and the design of search systems.},
  keywords = {Research Data Reuse, Humanities Research Data, Data Journals}
}

@inproceedings{fruth2026_115320,
  title = {ADISS : Authority Data Integration Search System},
  author = {Fruth, Leon and Gradl, Tobias and Henrich, Andreas},
  year = {2026},
  pages = {189-200},
  publisher = {Otto-Friedrich-Universität},
  address = {Bamberg},
  url = {https://fis.uni-bamberg.de/handle/uniba/115320},
  language = {eng},
  abstract = {This paper introduces ADISS, a generic search system designed to integrate heterogeneous authority file providers. Authority data is used to unambiguously identify entities such as persons, places, and organizations. As single data providers do not offer both quantity and quality of data, a combined access to multiple datasets is often required to support real-world use cases. In the context of Digital Humanities this combination improves the resolution of ambiguities in data curation processes. Our work is mainly motivated by two projects that require semi-automatic retrieval, as well as user-centered search scenarios for different authority file providers. Instead of using multiple existing endpoints to access the various datasets, we gather the heterogeneous data and make it accessible via integrated query and result models. In this paper, we present our highly configurable search API, which offers a diverse range of search and filtering options. We show that by its generic and highly configurable nature, our system is adaptable and reusable for a diverse set of use cases and conclude the paper with ideas for further steps and improvements.},
  keywords = {FAIR, Digital Humanities, Authority Data, Data Curation}
}

@inproceedings{hebeis2026_114779,
  title = {Automatisierte Typklassifikation von Normdaten mit BERT : Poster},
  author = {Hebeis, Maximilian and Fruth, Leon and Gradl, Tobias and Henrich, Andreas},
  year = {2026},
  publisher = {Otto-Friedrich-Universität},
  address = {Bamberg},
  urn = {urn:nbn:de:bvb:473-irb-114779x},
  url = {https://fis.uni-bamberg.de/handle/uniba/114779},
  language = {deu},
  abstract = {Normdaten sind zentral für die Interoperabilität geisteswissenschaftlicher Daten. Die an der Otto-Friedrich-Universität Bamberg entwickelte integrierte Suchplattform ADISS integriert verschiedene Normdatenquellen wie GND, Wikidata oder Geonames, jedoch bislang ohne einheitliche Typisierung. Dies erschwert eine gezielte Facettierung bei der Suche. Ziel des Projekts ist daher ein System zur automatisierten Typenzuordnung mittels maschinellen Lernens. Auf Basis von BERT soll ein Klassifikator trainiert werden, der Normdatensätze in ein reduziertes Zielschema überführt. Trainingsdaten entstehen durch Mapping zwischen GND- und Wikidata-Datensätzen. Neben Quelltypen werden Name und Beschreibung des jeweiligen Normdatensatzes als semantische Features genutzt. Erste Experimente mit einem auf einem multilingualen BERT-Modell als Encoder basierenden hierarchischen Klassifikator zeigen vielversprechende Ergebnisse. Künftig sollen die Datenbasis und Modellparameter optimiert werden, um eine robuste, domänenübergreifende Typisierung zur Verbesserung der Suchfunktionalität in ADISS zu ermöglichen.},
  keywords = {Normdaten, Textklassifikation, Maschinelles Lernen, Forschungsdateninfrastrukturen, Metadaten, Entity Typing, BERT, Datenintegration}
}

@inproceedings{hebeis2026_114771,
  title = {Automatisierte Typklassifikation von Normdaten mit BERT},
  author = {Hebeis, Maximilian and Fruth, Leon and Gradl, Tobias and Henrich, Andreas},
  year = {2026},
  pages = {447-450},
  publisher = {Otto-Friedrich-Universität},
  address = {Bamberg},
  urn = {urn:nbn:de:bvb:473-irb-114771x},
  url = {https://fis.uni-bamberg.de/handle/uniba/114771},
  language = {deu},
  abstract = {Normdaten sind zentral für die Interoperabilität geisteswissenschaftlicher Daten. Die an der Otto-Friedrich-Universität Bamberg entwickelte integrierte Suchplattform ADISS integriert verschiedene Normdatenquellen wie GND, Wikidata oder Geonames, jedoch bislang ohne einheitliche Typisierung. Dies erschwert eine gezielte Facettierung bei der Suche. Ziel des Projekts ist daher ein System zur automatisierten Typenzuordnung mittels maschinellen Lernens. Auf Basis von BERT soll ein Klassifikator trainiert werden, der Normdatensätze in ein reduziertes Zielschema überführt. Trainingsdaten entstehen durch Mapping zwischen GND- und Wikidata-Datensätzen. Neben Quelltypen werden Name und Beschreibung des jeweiligen Normdatensatzes als semantische Features genutzt. Erste Experimente mit einem auf einem multilingualen BERT-Modell als Encoder basierenden hierarchischen Klassifikator zeigen vielversprechende Ergebnisse. Künftig sollen die Datenbasis und Modellparameter optimiert werden, um eine robuste, domänenübergreifende Typisierung zur Verbesserung der Suchfunktionalität in ADISS zu ermöglichen.},
  keywords = {Normdaten, Textklassifikation, Maschinelles Lernen, Forschungsdateninfrastrukturen, Metadaten, Entity Typing, BERT, Datenintegration}
}

@inproceedings{hebeis2026_114773,
  title = {Automatisierte Typklassifikation von Normdaten mit BERT},
  author = {Hebeis, Maximilian and Fruth, Leon and Gradl, Tobias and Henrich, Andreas},
  year = {2026},
  booktitle = {Book of Abstracts : DHd 2026},
  pages = {447-450},
  publisher = {Zenodo},
  doi = {10.5281/zenodo.18703020},
  url = {https://fis.uni-bamberg.de/handle/uniba/114773},
  language = {deu},
  abstract = {Normdaten sind zentral für die Interoperabilität geisteswissenschaftlicher Daten. Die an der Otto-Friedrich-Universität Bamberg entwickelte integrierte Suchplattform ADISS integriert verschiedene Normdatenquellen wie GND, Wikidata oder Geonames, jedoch bislang ohne einheitliche Typisierung. Dies erschwert eine gezielte Facettierung bei der Suche. Ziel des Projekts ist daher ein System zur automatisierten Typenzuordnung mittels maschinellen Lernens. Auf Basis von BERT soll ein Klassifikator trainiert werden, der Normdatensätze in ein reduziertes Zielschema überführt. Trainingsdaten entstehen durch Mapping zwischen GND- und Wikidata-Datensätzen. Neben Quelltypen werden Name und Beschreibung des jeweiligen Normdatensatzes als semantische Features genutzt. Erste Experimente mit einem auf einem multilingualen BERT-Modell als Encoder basierenden hierarchischen Klassifikator zeigen vielversprechende Ergebnisse. Künftig sollen die Datenbasis und Modellparameter optimiert werden, um eine robuste, domänenübergreifende Typisierung zur Verbesserung der Suchfunktionalität in ADISS zu ermöglichen.},
  keywords = {Normdaten, Textklassifikation, Maschinelles Lernen, Forschungsdateninfrastrukturen, Metadaten, Entity Typing, BERT, Datenintegration}
}

@inproceedings{hebeis2026_114772,
  title = {Automatisierte Typklassifikation von Normdaten mit BERT : Poster},
  author = {Hebeis, Maximilian and Fruth, Leon and Gradl, Tobias and Henrich, Andreas},
  year = {2026},
  publisher = {Zenodo},
  doi = {10.5281/zenodo.18999712},
  url = {https://fis.uni-bamberg.de/handle/uniba/114772},
  language = {deu},
  abstract = {Normdaten sind zentral für die Interoperabilität geisteswissenschaftlicher Daten. Die an der Otto-Friedrich-Universität Bamberg entwickelte integrierte Suchplattform ADISS integriert verschiedene Normdatenquellen wie GND, Wikidata oder Geonames, jedoch bislang ohne einheitliche Typisierung. Dies erschwert eine gezielte Facettierung bei der Suche. Ziel des Projekts ist daher ein System zur automatisierten Typenzuordnung mittels maschinellen Lernens. Auf Basis von BERT soll ein Klassifikator trainiert werden, der Normdatensätze in ein reduziertes Zielschema überführt. Trainingsdaten entstehen durch Mapping zwischen GND- und Wikidata-Datensätzen. Neben Quelltypen werden Name und Beschreibung des jeweiligen Normdatensatzes als semantische Features genutzt. Erste Experimente mit einem auf einem multilingualen BERT-Modell als Encoder basierenden hierarchischen Klassifikator zeigen vielversprechende Ergebnisse. Künftig sollen die Datenbasis und Modellparameter optimiert werden, um eine robuste, domänenübergreifende Typisierung zur Verbesserung der Suchfunktionalität in ADISS zu ermöglichen.},
  keywords = {Normdaten, Textklassifikation, Maschinelles Lernen, Forschungsdateninfrastrukturen, Metadaten, Entity Typing, BERT, Datenintegration}
}

@inproceedings{fruth2026_114695,
  title = {Cataloguing Editions and Other Resources in One Unified System : The Case of the Text+ Registry},
  author = {Fruth, Leon and Geißler, Nils and Gradl, Tobias and Schulz, Daniela},
  year = {2026},
  publisher = {Otto-Friedrich-Universität},
  address = {Bamberg},
  urn = {urn:nbn:de:bvb:473-irb-114695x},
  url = {https://fis.uni-bamberg.de/handle/uniba/114695},
  language = {eng},
  abstract = {Text+ is a consortium of the German National Research Data Infrastructure (Nationale Forschungsdateninfrastruktur, NFDI). With an initial focus on the three data domains collections, lexical resources, and editions, Text+ aims at constructing a research data infrastructure that is both flexible and scalable, catering to the unique requirements of academic disciplines that work with textual and linguistic data, but also beyond. Its goal is to further develop standards and best practices to make data available according to the FAIR principles (Findable, Accessible, Interoperable, Reusable). As an initiative of the community for the community, the consortium provides a whole range of services, data and consulting for institutions, projects and individual researchers. 

One pivotal component within the Text+ infrastructure is the Text+ Registry, which is designed as a unified cataloguing system for the diverse set of resources from the data domains. The Registry builds on data that has been gathered and made accessible by existing databases and catalogues (like Greta Franzini’s Dig-Ed-Cat and Patrick Sahle’s Catalog of Digital Scholarly Editions, library or union catalogues, research information systems of funding bodies such as AGATE). An often lacking integration with external knowledge bases and contextual biases (discipline, type of resource) in external sources are resolved by compiling descriptions from multiple sources wherever available and possible. Remaining gaps are filled in a manual curation layer by respective domain experts. In order to be able to accommodate the diverse resource landscape, the Registry is technologically based on a sophisticated metamodelling concept that allows the definition and evolution of dedicated data models for different types of resources, and enables the import, curation and provision of resource descriptions. APIs for data ingest from existing catalogues and data provision are provided, and search and retrieval tools such as the CLARIN Federated Content Search (FCS) and the DARIAH-DE Generic Search (GS) build on these interfaces. The Text+ Registry should not only increase the visibility of scientific resources, but also overcome the traditional barriers that hinder their discoverability and accessibility.

The data models in the Text+ Registry are tailored to different requirements for each data domain and were developed based on relevant subject-specific standards and best practices, as well as in direct collaboration with the communities (e.g. Specialised Information Services, SIS). The formalisation of relationships between collections, editions and lexical resources, as well as tools, actors and other entity types, creates a resource network that can also be expanded over and above Text+ to prepare for integration into further developments within the NFDI context such as knowledge graphs.

The paper presents the technologies and methodologies behind the Text+ Registry and reflects on challenges, opportunities but also limitations. Taking the data model for cataloguing editions as an example, the metamodelling approach as well as the architectural blueprint of the Text+ Registry are presented.},
  keywords = {Text+, Data Integration, Metadata Catalogues, Digital Humanities}
}

@mastersthesis{fruth2026_112994,
  title = {An Approach Towards Unsupervised Text Simplification on Paragraph-Level for German Texts},
  author = {Fruth, Leon},
  year = {2026},
  publisher = {Otto-Friedrich-Universität},
  address = {Bamberg},
  doi = {10.20378/irb-112994},
  urn = {urn:nbn:de:bvb:473-irb-112994x},
  url = {https://fis.uni-bamberg.de/handle/uniba/112994},
  language = {eng},
  abstract = {Text simplification aims to make texts easier to read and comprehend for people. Recent approaches tackle this task by training neural models on large-scale parallel datasets. However, most languages only have limited simplification data available. Laban et al. [2021b] presented an unsupervised simplification approach to avoid the need for parallel data. They introduced the training algorithm k-SCST, which optimizes a reward by generating and scoring multiple candidate simplifications and encouraging the candidates that outperform the mean reward. This thesis adapts this approach to simplify short paragraphs from German Wikipedia articles. The individual scores of the reward regarding simplicity, fluency and meaning preservation are modified for the new domain and language. In addition, other aspects of the training method are explored. The results show some lexical and syntactic simplification phenomena but also problems regarding fluency and faithfulness. The findings are assessed, and suggestions for future improvements are presented.},
  keywords = {Text Simplification, Natural Language Processing, Unsupervised Machine Learning, German Text Processing, Language Models},
  school = {Otto-Friedrich-Universität Bamberg}
}

@article{fruth2026_112845,
  title = {A Flexible Search System for Integrated Authority Data : ADISS},
  author = {Fruth, Leon and Gradl, Tobias and Hebeis, Maximilian and Henrich, Andreas},
  year = {2026},
  pages = {167-178},
  publisher = {Otto-Friedrich-Universität},
  address = {Bamberg},
  urn = {urn:nbn:de:bvb:473-irb-112845x},
  url = {https://fis.uni-bamberg.de/handle/uniba/112845},
  language = {eng},
  abstract = {This paper presents ADISS, a generic and flexible search system designed to integrate heterogeneous authority file providers. Authority data is used to unambiguously identify entities such as persons, places, and institutions. As individual data providers typically do not offer both broad, universal coverage and domain-specific, in-depth data, many real-world application scenarios require combined access to multiple datasets. In the context of Digital Humanities, combining multiple authority data providers helps to improve the coverage and to resolve ambiguities more effectively during data curation. Instead of relying on separate endpoints to access multiple data sources, we address this need by aggregating the datasets and exposing them through integrated query and result models, thus simplifying access at the level of technical interfaces and schema integration. We present a highly configurable search API, which offers a diverse range of search and filtering options, to leverage the integrated data sources. We demonstrate its practical value through two active projects, that require semi-automatic retrieval, as well as user-centered search scenarios for different authority file providers. An evaluation based on manually curated data shows high retrieval accuracy, highlighting the system’s effectiveness. We further illustrate how ADISS’s generic and highly configurable design makes it adaptable and reusable for a wide range of use cases, and we outline directions for future improvements.},
  keywords = {FAIR, Digital Humanities, Authority Data, Data Curation}
}

@inproceedings{fruth2025_114701,
  title = {Cataloguing Editions and Other Resources in One Unified System : The Case of the Text+ Registry},
  author = {Fruth, Leon and Geißler, Nils and Gradl, Tobias and Schulz, Daniela},
  year = {2025},
  booktitle = {Proceedings of the Digital Humanities Congress 2024},
  publisher = {The Digital Humanities Institute (DHI)},
  address = {Sheffield},
  url = {https://fis.uni-bamberg.de/handle/uniba/114701},
  language = {eng},
  abstract = {Text+ is a consortium of the German National Research Data Infrastructure (Nationale Forschungsdateninfrastruktur, NFDI). With an initial focus on the three data domains collections, lexical resources, and editions, Text+ aims at constructing a research data infrastructure that is both flexible and scalable, catering to the unique requirements of academic disciplines that work with textual and linguistic data, but also beyond. Its goal is to further develop standards and best practices to make data available according to the FAIR principles (Findable, Accessible, Interoperable, Reusable). As an initiative of the community for the community, the consortium provides a whole range of services, data and consulting for institutions, projects and individual researchers. 

One pivotal component within the Text+ infrastructure is the Text+ Registry, which is designed as a unified cataloguing system for the diverse set of resources from the data domains. The Registry builds on data that has been gathered and made accessible by existing databases and catalogues (like Greta Franzini’s Dig-Ed-Cat and Patrick Sahle’s Catalog of Digital Scholarly Editions, library or union catalogues, research information systems of funding bodies such as AGATE). An often lacking integration with external knowledge bases and contextual biases (discipline, type of resource) in external sources are resolved by compiling descriptions from multiple sources wherever available and possible. Remaining gaps are filled in a manual curation layer by respective domain experts. In order to be able to accommodate the diverse resource landscape, the Registry is technologically based on a sophisticated metamodelling concept that allows the definition and evolution of dedicated data models for different types of resources, and enables the import, curation and provision of resource descriptions. APIs for data ingest from existing catalogues and data provision are provided, and search and retrieval tools such as the CLARIN Federated Content Search (FCS) and the DARIAH-DE Generic Search (GS) build on these interfaces. The Text+ Registry should not only increase the visibility of scientific resources, but also overcome the traditional barriers that hinder their discoverability and accessibility.

The data models in the Text+ Registry are tailored to different requirements for each data domain and were developed based on relevant subject-specific standards and best practices, as well as in direct collaboration with the communities (e.g. Specialised Information Services, SIS). The formalisation of relationships between collections, editions and lexical resources, as well as tools, actors and other entity types, creates a resource network that can also be expanded over and above Text+ to prepare for integration into further developments within the NFDI context such as knowledge graphs.

The paper presents the technologies and methodologies behind the Text+ Registry and reflects on challenges, opportunities but also limitations. Taking the data model for cataloguing editions as an example, the metamodelling approach as well as the architectural blueprint of the Text+ Registry are presented.},
  keywords = {Text+, Data Integration, Metadata Catalogues, Digital Humanities}
}

@inproceedings{gradl2025_112303,
  title = {Teilprojekt Text+ – Nutzbarmachung von Sammlungen, Editionen und Co},
  author = {Gradl, Tobias and Hirmer, Tobias and Fruth, Leon and Henrich, Andreas},
  year = {2025},
  publisher = {Otto-Friedrich-Universität},
  address = {Bamberg},
  doi = {10.20378/irb-112303},
  urn = {urn:nbn:de:bvb:473-irb-112303x},
  url = {https://fis.uni-bamberg.de/handle/uniba/112303},
  language = {deu},
  abstract = {Text+ ist ein Konsortium der Nationalen Forschungsdateninfrastruktur (NFDI) und bietet Zugang zu Knowhow, Standards und Services im Bereich der text- und sprachbasierten Wissenschaften. In Text+ ist das Team des Lehrstuhls Medieninformatik insbesondere mit Themen rund um die Erschließung und Nutzbarmachung von Datenbeständen befasst. Dabei werden die Anbieter von Forschungsdaten, die sogenannten Datenzentren, bei der Umsetzung der FAIR Prinzipien unterstützt. Über die anbieterorientierte Perspektive hinaus stehen Anforderungen durch die Fachwissenschaften in der Forschung im Vordergrund. Anforderungen aus DiSo an die über Text+ zugänglichen Ressourcen können für die Gestaltung von Diensten und Standards mit Blick auf die Nachnutzbarkeit in Unterricht und Lehrkräftebildung eingebracht und berücksichtigt werden.},
  keywords = {DiSo, lernen:digital}
}

@inproceedings{fruth2025_111935,
  title = {An Approach towards Unsupervised Text Simplification on Paragraph-Level for German Texts},
  author = {Fruth, Leon and Jegan, Robin and Henrich, Andreas},
  year = {2025},
  pages = {77–89},
  publisher = {Otto-Friedrich-Universität},
  address = {Bamberg},
  urn = {urn:nbn:de:bvb:473-irb-111935x},
  url = {https://fis.uni-bamberg.de/handle/uniba/111935},
  language = {eng},
  abstract = {Text simplification as a research field has received attention in recent years for English and other languages, however, German text simplification techniques are lacking thus far. We present an unsupervised simplification approach for German texts using reinforcement learning (self-critical sequence training). Our main contributions are the adaption of an existing method for English, the selection and creation of German corpora for this task and the customization of rewards for particular aspects of the German language. In our paper, we describe our system and an evaluation, including still present issues and problems due to the complexity of the German language, as well as directions for future research.},
  keywords = {Unsupervised Text Simplification}
}

@article{fruth2025_111276,
  title = {A Flexible Search System for Integrated Authority Data : ADISS},
  author = {Fruth, Leon and Gradl, Tobias and Hebeis, Maximilian and Henrich, Andreas},
  year = {2025},
  journal = {Datenbank-Spektrum : Zeitschrift für Datenbanktechnologie ; Organ der Fachgruppe Datenbanken der Gesellschaft für Informatik e.V.},
  volume = {25},
  number = {3},
  pages = {167-178},
  publisher = {Springer},
  address = {Berlin ; Heidelberg},
  issn = {1618-2162},
  doi = {10.1007/s13222-025-00515-7},
  url = {https://fis.uni-bamberg.de/handle/uniba/111276},
  language = {eng},
  abstract = {This paper presents ADISS, a generic and flexible search system designed to integrate heterogeneous authority file providers. Authority data is used to unambiguously identify entities such as persons, places, and institutions. As individual data providers typically do not offer both broad, universal coverage and domain-specific, in-depth data, many real-world application scenarios require combined access to multiple datasets. In the context of Digital Humanities, combining multiple authority data providers helps to improve the coverage and to resolve ambiguities more effectively during data curation. Instead of relying on separate endpoints to access multiple data sources, we address this need by aggregating the datasets and exposing them through integrated query and result models, thus simplifying access at the level of technical interfaces and schema integration. We present a highly configurable search API, which offers a diverse range of search and filtering options, to leverage the integrated data sources. We demonstrate its practical value through two active projects, that require semi-automatic retrieval, as well as user-centered search scenarios for different authority file providers. An evaluation based on manually curated data shows high retrieval accuracy, highlighting the system’s effectiveness. We further illustrate how ADISS’s generic and highly configurable design makes it adaptable and reusable for a wide range of use cases, and we outline directions for future improvements.},
  keywords = {FAIR, Digital Humanities, Authority Data, Data Curation}
}

@inproceedings{gradl2025_111116,
  title = {The Text+ Registry : Federating Research Data Catalogues for the Digital Humanities},
  author = {Gradl, Tobias and Fruth, Leon and Henrich, Andreas},
  year = {2025},
  booktitle = {Linking Theory and Practice of Digital Libraries : 29th International Conference on Theory and Practice of Digital Libraries, TPDL 2025, Tampere, Finland, September 23–26, 2025, Proceedings},
  pages = {378-394},
  publisher = {Springer Nature Switzerland},
  address = {Cham},
  isbn = {978-3-032-05409-8},
  doi = {10.1007/978-3-032-05409-8\_22},
  url = {https://fis.uni-bamberg.de/handle/uniba/111116},
  language = {eng}
}

@inproceedings{fruth2025_108435,
  title = {ADISS : Authority Data Integration Search System},
  author = {Fruth, Leon and Gradl, Tobias and Henrich, Andreas},
  editor = {Gesellschaft für Informatik, Bonn},
  year = {2025},
  booktitle = {Datenbanksysteme für Business, Technologie und Web : Workshopband (BTW 2025)},
  pages = {189-200},
  publisher = {Gesellschaft für Informatik},
  address = {Bonn},
  doi = {10.18420/btw2025-123},
  url = {https://fis.uni-bamberg.de/handle/uniba/108435},
  language = {eng},
  abstract = {This paper introduces ADISS, a generic search system designed to integrate heterogeneous authority file providers. Authority data is used to unambiguously identify entities such as persons, places, and organizations. As single data providers do not offer both quantity and quality of data, a combined access to multiple datasets is often required to support real-world use cases. In the context of Digital Humanities this combination improves the resolution of ambiguities in data curation processes. Our work is mainly motivated by two projects that require semi-automatic retrieval, as well as user-centered search scenarios for different authority file providers. Instead of using multiple existing endpoints to access the various datasets, we gather the heterogeneous data and make it accessible via integrated query and result models. In this paper, we present our highly configurable search API, which offers a diverse range of search and filtering options. We show that by its generic and highly configurable nature, our system is adaptable and reusable for a diverse set of use cases and conclude the paper with ideas for further steps and improvements.},
  keywords = {FAIR, Digital Humanities, Authority Data, Data Curation}
}

@inproceedings{fruth2024_103945,
  title = {An Approach towards Unsupervised Text Simplification on Paragraph-Level for German Texts},
  author = {Fruth, Leon and Jegan, Robin and Henrich, Andreas},
  year = {2024},
  booktitle = {Proceedings of the Workshop on DeTermIt! Evaluating Text Difficulty in a Multilingual Context @ LREC-COLING 2024},
  pages = {77–89},
  publisher = {ELRA and ICCL},
  url = {https://fis.uni-bamberg.de/handle/uniba/103945},
  language = {eng},
  abstract = {Text simplification as a research field has received attention in recent years for English and other languages, however, German text simplification techniques are lacking thus far. We present an unsupervised simplification approach for German texts using reinforcement learning (self-critical sequence training). Our main contributions are the adaption of an existing method for English, the selection and creation of German corpora for this task and the customization of rewards for particular aspects of the German language. In our paper, we describe our system and an evaluation, including still present issues and problems due to the complexity of the German language, as well as directions for future research.},
  keywords = {Unsupervised Text Simplification}
}

@inproceedings{jegan2023_59341,
  title = {Integrating Access to Authority Data for Improved Interoperability of Research Data in the Digital Humanities},
  author = {Jegan, Robin and Fruth, Leon and Gradl, Tobias and Henrich, Andreas},
  year = {2023},
  pages = {829-836},
  publisher = {Otto-Friedrich-Universität},
  address = {Bamberg},
  urn = {urn:nbn:de:bvb:473-irb-593413},
  url = {https://fis.uni-bamberg.de/handle/uniba/59341},
  language = {eng},
  abstract = {Authority data is used to unambiguously identify persons, organisations and places. In this paper, a means to integrate access to several providers of authority data into data curation processes is described, which facilitates disambiguation of geographic data. Combined access to general datasets, in our case the Gemeinsame Normdatei (GND), as well as highly specialized datasets, here the Memorial Archives, improves the resolution of ambiguities and particularly benefits use cases of the Digital Humanities. The integration is necessary in order to abstract from technical, syntactical and semantic heterogeneity of the providers. Operations such as querying geographic information and receiving enriched data from different data sources are facilitated. An overview of the goals of the system, related projects and authority data providers are presented, as well as details on the implementation and further steps.},
  keywords = {authority file, geographic data, bases data integration}
}

@inproceedings{jegan2023_59235,
  title = {Integrating Access to Authority Data for Improved Interoperability of Research Data in the Digital Humanities},
  author = {Jegan, Robin and Fruth, Leon and Gradl, Tobias and Henrich, Andreas},
  year = {2023},
  booktitle = {Datenbanksysteme für Business, Technologie und Web (BTW 2023) : 06.-10. März 2023 in Dresden, Deutschland},
  pages = {829-836},
  publisher = {Gesellschaft für Informatik e.V.},
  isbn = {978-3-88579-725-8},
  doi = {10.18420/BTW2023-54},
  url = {https://fis.uni-bamberg.de/handle/uniba/59235},
  language = {eng},
  abstract = {Authority data is used to unambiguously identify persons, organisations and places. In this paper, a means to integrate access to several providers of authority data into data curation processes is described, which facilitates disambiguation of geographic data. Combined access to general datasets, in our case the Gemeinsame Normdatei (GND), as well as highly specialized datasets, here the Memorial Archives, improves the resolution of ambiguities and particularly benefits use cases of the Digital Humanities. The integration is necessary in order to abstract from technical, syntactical and semantic heterogeneity of the providers. Operations such as querying geographic information and receiving enriched data from different data sources are facilitated. An overview of the goals of the system, related projects and authority data providers are presented, as well as details on the implementation and further steps.},
  keywords = {authority file, geographic data, bases data integration}
}

