The third edition of the Biographical Data in a Digital World Conference took place on 5 and 6 September 2019 in Varna, Bulgaria. The conference included nine long and four short presentations and a session with work groups on modeling biographical data. These proceedings include nine full papers that were accepted upon single blind review. The first two editions of Biographical Data in a Digital World provided a, for many researchers, first opportunity to connect with other researchers working on digital biographical resources. They brought together a wide variety of perspectives from historians, librarians, literature studies, computer scientists and computational linguistics. Next to differences and new angles, we also found commonalities: a shared interest in the richness and variety of the resources, challenges with gaps in the data and data quality, approaches to data representation and horizons waiting to be explored on several levels. Projects in various countries are steadily making progress, resources are growing and methods for exploring them are improving. At an international level, we find a continuous wish to connect. As a community, we want to exchange ideas and learn from each other, but in particular, we are interested in identifying connections among our resources. This desire forms the main motivation for continuing to organize events in the BD-conference series, which resulted in the third edition of this conference and (with a long partially Covid-19 related delay) these proceedings.
The third edition of Biographical Data in a Digital World
took place in Varna, B
Because interdisciplinary venues bring together different publication cultures, researchers from various domains in digital humanities, may not be familiar with the workshop and conference proceedings publication culture common in computer science. We therefore opted for three modes of acceptance for full papers: regular acceptance (with a chance to make updates and encouragement to incorporate feedback for the camera ready version), acceptance with minor revisions (with a requirement to address the main criticism of reviewers which would be verified by the editors) and major revisions (with an independent review round of the revised paper which may lead to a decision that the work cannot be included in the proceedings). For all these modes of acceptance, authors could come and present their work. Camera ready versions were collected after the event, so that authors could also make use of feedback from other participants.
Four papersB received a regular accept, three were accepted conditionally based on minor revisions, two papers received a major revision request and one was rejected. All vfie abstracts were deemed relevant and accepted for presentation (without publication). With one author not being able to make it, there were thirteen presentations around various topics at the event. All accepted papers submitted revised versions. The minor revision papers were checked by the editorial board, which also formed a new reviewing committee for the two remaining papers. One needed an extra round of minor revisions after which all completed papers could be included in the proceedings. An overview of the papers is given below.
In addition to the work presented in this paper, various discussions around connecting resources from different countries took place during the event. This resulted in the proposal of the Horizon2020 project InTaVia1 that is currently running bringing together researchers from Austria, Denmark, Germany, the Netherlands and Slovenia who met through the Biographical Data conference series. 3
The papers in these proceedings cover three themes i) Building Digital Biographical Resources ii) applying NLP tools for biographical data mining, and iii) Exploring capabilities of digital resources with case studies.
Bhreathnach et al. (2019) identify existing biases, features and omissions in the Irish biographical database Ainm.2 This is an analysis on the distribution of biographies in terms of people’s lifespans, gender and birthplace, as well as professions present in the mentioned resource. They 1https://intavia.eu 2https://www.ainm.ie/Info.aspx?Topic=resources.en found, for example, that even though the resource spans from 15th to 21st centuries, the database has a heavy bias towards people living in the 19th and 20th century (around 75% of the subjects). This work also quantitatively conifrms an already acknowledged bias towards people from certain regions and professions.
Hyvo¨nen et al. (2019) argue for shifting towards Linked Data as the paradigm for publishing and using biographical dictionaries. To support their arguments, they describe the data service and semantic portal BiographySampo,3 where biography texts are enriched with 16 external data sources. Their paper outlines how reasoning through the structured graph can be used to expand and discover serendipitious relations among entities, persons, and places.
Povroznik (2019) studies the collective portraits of deputies of local self-government in Russia in the second half of the XIX century. With this goal in mind, the paper describes the problems of searching, organizing, modelling, analyzing and presenting data during the process for building a useful database for prosopographical research.
Vogeler et al. (2019) discuss the viability of an international prosopographical framework. The authors propose a set of data resources, interfaces and analytical tools to integrate and facilitate the access to prosopographical resources. Their solution is a data model that can be accessed by a RESTful API. The paper illustrates this idea through a concrete example based on resources built for the Austrian Prosopographical database (APIS) by following their proposed methods.
Magistry et al. (2019) use Natural Language Processing (NLP) tools to extract information and create a structured data resource based on the biographical Dictionary of Republican China (BDRC). Their approach also uses the constructed graphs as means for exploration on the relationships centered on education and position occupied by the people on the database.
Pl
Filipov et al. (2019) present a visualization tool for interactive analysis on multiple biographical timelines, relating different biographies through a specific set of events and locations. Specifically, they show the example of biographies connected to Austrian music history. Through their visualization technique, new potential narratives can be created and contextualized, providing an extra layer on the process of historical research.
Koho et al. (2019) introduce a database of Finnish prisoners of war in the Soviet Union. The project is targeted to the
researchers of military history. It aims at centralizing all information available from different data sources and linking them to each prisoner. This database facilitates research that explores the details, relations and subgroups of each individual biography.
Mayr et al. (2019) describe research that is part of the Polycube project. Polycube’s goal is to visualize multiple data dimensions such as space, categories and relations over time. This paper particularly raises the question of how visualizations of life and work go together? They answer it with the case study of Charles W. Cushman showing how these visual-analytical frames of reference could provide a multimodal, narrative framework of biographical knowledge exploration and communication. We would like to thank the program committee members for their careful and critical reviews, which supported the selection process and helped authors improve their papers. We would also like to thank the other members of the editorial board for their input during discussions and the careful checks and reviews for the conditionally accepted papers. Many special thanks go to the local organizers: Petya Osenova, Kiril Simov and Alexander Popov for an excellent job resulting in an impeccably organized conference. Finally, we would like to thank everyone involved for their patience and understanding for the delays in bringing out these proceedings.
Antske Fokkens, Vrije Universiteit Amsterdam & Eindhoven University of Technology Petya Osenova, Bulgarian Academy of Sciences Kiril Simov, Bulgarian Academy of Sciences Alexander Popov, Bulgarian Academy of Sciences Paul Arthur, Edith Cowan University Serge ter Braake, Stichting UvO Thierry Declerck, DFKI Saarland Ronald Sluijter, Huygens ING Eveline Wandl-Vogt, Austrian Academy of Sciences, Ars Electronica Research Institute knowledge for humanity
Serge ter Braake, Stichting UvO Thierry Declerck, DFKI Saarland Lonneke Geerlings, Stichting UvO Gernot Howanitz, Universita¨t Innsbruck Anders Ingram, University of Oxford Ba¨rbel Kro¨ger, Akademie der Wissenschaften zu Go¨ttingen Eetu Ma¨kela¨, Aalto University Lodewijk Petram, Huygens ING Katharina Prager, Ludwig Boltzmann Institute for Digital History Matthias Reinert, Neue Deutsche Biographie, Germany Matthias Schlo¨gl, O¨sterreichische Akademie der Wissenschaften Kiril Simov, Bulgarian Academy of Sciences Ronald Sluijter, Huygens ING Petra Vide Ogrin, Slovenian Academy of Sciences and Arts Georg Vogeler, O¨sterreichische Akademie der Wissenschaften Marcos Zampieri, Rochester Institute of Technology Kalliopi Zervanou, Leiden University Joris van Zundert, Huygens ING