Digital Database of WWI Victims from Slovenia (ZV1): Project Cooperation Between the Digital Humanities and Cultural Heritage

Various events, which were related to marking the 100th anniversary of WWI, have in recent years further increased the interest in historical resources and research on WWI (1914–1918). Some successful projects took place on the international level, for example, the Europeana Collections 1914–1918, 1914–1918-online and CENDARI, while even more were carried out at the national level. Most of these projects aim at securing access to the digitized cultural WWI heritage, which is kept by GLAM institutions (galleries, libraries, archives, and museums). Some institutions also published databases of the (fallen) soldiers from the WW1. (Clavert, 2018a, 2018b) The poster will present a digital database of WWI victims from Slovenia, which was created in cooperation between various research and cultural institutions. As many as 10 museums, two archive facilities and two research institutions participated in the consortium. The project was coordinated by the Institute of Contemporary History, which has a well-developed Digital Humanities Research Group, and acts as a national coordinating institution for DARIAH in Slovenia.
The project of collecting data on the WWI victims from Slovenia (
ZV1) is a good example of successful institutional integration of the local GLAM institutions, research organizations and the digital research infrastructure. Since the project was created upon the initiative of the GLAM Institutions, it provides a good example of successfully overcoming the challenges that the initiators of the Cultural Heritage Data Reuse Charter are facing. (Seillier, 2018)

The work of the consortium partners was based on the following principles:

Separate funding: The project was financed from the existing activities of individual partners.

Use of the existing databases: Partners contributed data from their existing databases.

Separate data collection: Partners collected data primarily for the purpose of activities carried out in the parent institutions.

Open access: The collected information is made available by the partners under the terms of the Creative Commons licenses.

Uniform data protection: Partners stored the collected data in the verified repository.

Single database: Collected data from all partners were unified and imported into the MySQL relational database.

We opted for this type of operation for the following reasons:

For the purpose of collecting data on all WWI victims, it was impossible to obtain adequate project financing.
Although a lot of volunteer work was carried out as part of the project, the intensity of the project work was determined by the partners alone, according to their wishes and expectations.
The consortium has brought together researchers from diverse research institutions, who had varying levels of technical knowledge and experience. This way, the project partners selected the appropriate tools and software by themselves. Most of them used spreadsheet programs, primarily the Microsoft Excel.

The designed project had the following shortcomings:

Missing data: Since the project partners mostly collected data for a specific region and the consortium was not joined by all relevant institutions from Slovenia, in some regions the data was not collected at all.

Non-uniform data: Since the project partners collected data for various reasons, the collected data was not always interoperable.

Duplicates: Due to the fact that data was collected by different partners, whereby each did it for their own purpose, it is possible that multiple records were collected for the same person within the common database.

The Digital Humanities Research Group eliminated these deficiencies in the following ways:

Missing data: Following the model of the German and Austrian genealogical projects, the group used the digitized Casualty Lists (Verlustliste) for the entire Austria-Hungary. On the
ANNO website, we have scraped OCRs of all available casualty lists in the TXT format. When converting from the TXT format to TEI XML, we used the semi-automatic annotation: regular expression, XPath, XSL Stylesheets (
GitHub). We used the TEI Module for Names, Dates, People, and Places (Figure 1) (TEI Consortium, 2018).

Non-uniform data: The ability to import and export data in the CSV format and subsequent cleaning of messy data: OpenRefine, R.

Duplicates: The core record linkage and deduplication tool is already available in the ZV1 database. For more advanced deduplication of people records, we use the RecordLinkage R-package and the Fine-grained Record Integration and Linkage Tool (FRIL).

An example of an XML TEI markup for WWI victims

The editorial project team established a unified data model on the basis of which a MySQL relational database was built. There are currently 26.205 people in the database. The digital base of the WWI victims was designed by using traditional online technologies: HTML5, CSS, PHP7 and JavaScript. We use the ElasticSearch engine to perform the base search. For a tabular display of search results, we use the DataTables, which enables further filtering of data by selected columns (Figure 2).

ZV1 database

Future Plans:

Short-term: To convert as many source databases from project partners into TEI as possible and publish them as digital editions.

Long-term: Within the project, GeoNames was used to mark historic places. At the same time, we are planning to classify events and military units. Following the example of successful Linked Open Data (LOD) projects that used WWI data (Mäkelä et al., 2017), our long-term goal is to publish information about the WWI victims from Slovenia on the semantic web.

Bibliography

Clavert, F. (2018a).
Traces de mémoires collectives et bases de données de la Première Guerre mondiale. [Blog] L'histoire contemporaine à l'ère numérique. https://histnum.hypotheses.org/2743 [Accessed 4 April 2019].

Clavert, F. (2018b). WW1 (fallen) soldiers on-line: databases as traces of collective memories.
Digital Humanities Benelux Conference, Amsterdam, June 2018.

Mäkelä, E., Törnroos, J., Lindquist, T. and Hyvönen, E. (2017). WW1LOD – An application of CIDOC-CRM to World War 1 Linked Data.
International Journal on Digital Libraries, 18(4): 333–343.

Seillier, D. (2018).
Assessing the survey on the Charter’s principles. [Blog] Heritage Data Reuse Charter: Information and development around the Heritage Reuse Charter. https://datacharter.hypotheses.org/179 [Accessed 4 April 2019].

TEI Consortium (2018).
Guidelines for Electronic Text Encoding and Interchange. http://www.tei-c.org/P5/ [Accessed 4 April 2019].