Periodic Reporting for period 2 - InTaVia (In/Tangible European Heritage - Visual Analysis, Curation and Communication)
Reporting period: 2021-11-01 to 2023-10-31
A treasure trove of objects and texts in European museums, galleries, archives, and libraries documents the continent’s history from a variety of angles. However, countless connections between cultural objects and corresponding cultural (hi)stories have not been established at the current state of the digitization of Europe's cultural heritage. In this context, the InTaVia project drew together information from multiple national databases on intangible assets - such as biographical data about historical persons - and from databases on tangible cultural objects and linked them in a knowledge graph. Its overall aim was to develop an information portal for the integration, visual analysis, and visual communication of these assets. For this purpose, it developed new means of visually supported data creation, curation, analysis and storytelling. As such, the riches of European culture become more accessible globally - for experts and the public alike.
The InTaVia project developed a system architecture for the ingestion, linking and enrichment of data on cultural objects and related person biographies from different European cultural databases. The resulting InTaVia KNOWLEDGE GRAPH (IKG) contains 24,588,310 structured statements on 713,570 cultural objects and persons. To find further interesting links between these entities, we used AI technologies for knowledge discovery. To generate structured data from biographical texts (e.g. from biography databases, but also from Wikipedia) for their use in the IKG, tools for natural language processing (NLP) have been further developed, which can cope with their textual idiosyncrasies like abbreviations and language change.
To enhance access and human readability of this knowledge graph, we developed a WEB-BASED PLATFORM for the curation, analysis and communication of cultural data centered on methods of data visualization and storytelling (https://intavia.acdh-dev.oeaw.ac.at/ ). It consists of (1) a data curation lab (DCL) for searching, inspecting, curating, and compiling data from the IKG, but also from locally curated and imported data, (2) a visual analytics studio (VAS), for the visual analysis of geographical distributions and trajectories, temporal developments, and relations within data collections, and (3) a visual storytelling suite (STS) for creating stories composed of texts, visualizations and other media with a story creator (SC) and a corresponding story viewer (SV) to facilitate audiences’ reception on different devices.
As possible USERS and beneficiaries of the InTaVia platform, we defined 10 personas and various user practices, and involved representatives in three evaluation cycles to iteratively refine and improve the platform. Several tutorials were developed for guidance to the InTaVia platform. Five example stories have been published, which demonstrate the range of use cases of the storytelling suite.
For the DISSEMINATION and EXPLOITATION of results, we organized several events (7 workshops, 3 conferences) for different target audiences and published the project findings for multiple audiences and communities.
To enhance access and human readability of this knowledge graph, we developed a WEB-BASED PLATFORM for the curation, analysis and communication of cultural data centered on methods of data visualization and storytelling (https://intavia.acdh-dev.oeaw.ac.at/ ). It consists of (1) a data curation lab (DCL) for searching, inspecting, curating, and compiling data from the IKG, but also from locally curated and imported data, (2) a visual analytics studio (VAS), for the visual analysis of geographical distributions and trajectories, temporal developments, and relations within data collections, and (3) a visual storytelling suite (STS) for creating stories composed of texts, visualizations and other media with a story creator (SC) and a corresponding story viewer (SV) to facilitate audiences’ reception on different devices.
As possible USERS and beneficiaries of the InTaVia platform, we defined 10 personas and various user practices, and involved representatives in three evaluation cycles to iteratively refine and improve the platform. Several tutorials were developed for guidance to the InTaVia platform. Five example stories have been published, which demonstrate the range of use cases of the storytelling suite.
For the DISSEMINATION and EXPLOITATION of results, we organized several events (7 workshops, 3 conferences) for different target audiences and published the project findings for multiple audiences and communities.
InTaVia moved beyond the scientific and technological state of the art on several levels:
(1) Currently, cultural information is mostly stored in “silos”, i.e. isolated databases containing cultural heritage and history information. InTaVia developed methods and tools to interlink different databases - with different kinds of cultural assets, different languages, and different data models. We developed the InTaVia data model (IDM, building on CIDOC CRM) and reconciled information from different biographical and object databases with it.
(2) In cultural heritage research fields, a lot of information is stored in textual format. For their interlinking in knowledge graphs like the IKG, structured data has to be created from such texts. InTaVia further developed methods of natural language processing and artificial intelligence to create structured data and relations from textual information.
(3) Until now, interactive visual interfaces did focus either on persons or on cultural objects, but were not able to handle different kinds of entities. The visualizations developed within InTaVia represent objects, persons, but also places and organizations, in a coherent fashion and thereby enable their integrated visual analysis and communication.
(4) While some methods used for the development of theInTaVia are not new, the InTaVia platform is the first of its kind, which supports the full practitioner workflow from searching for information, to the creation of new data, the curation of erroneous data, its visual analysis, and its narrative communication.
(5) The InTaVia project consortium makes its results, methods, and prototypes fully and openly accessible for re-use by the scientific community, European research infrastructures, GLAM institutions, and cultural and creative industries. Open workflows and APIs allow the ingestion of further data to the IKG, the import of local data to the platform, and the re-use in other tools.
With these innovations, the InTaVia consortium expects these developments to have a notable and measurable impact on the local and global promotion of Europe's cultural heritage, on the analysis and understanding of cultural history, on future practices of curating digital assets, and on their use and re-use in various fields including Academia, education, cultural tourism, cultural journalism, and creative industries.
(1) Currently, cultural information is mostly stored in “silos”, i.e. isolated databases containing cultural heritage and history information. InTaVia developed methods and tools to interlink different databases - with different kinds of cultural assets, different languages, and different data models. We developed the InTaVia data model (IDM, building on CIDOC CRM) and reconciled information from different biographical and object databases with it.
(2) In cultural heritage research fields, a lot of information is stored in textual format. For their interlinking in knowledge graphs like the IKG, structured data has to be created from such texts. InTaVia further developed methods of natural language processing and artificial intelligence to create structured data and relations from textual information.
(3) Until now, interactive visual interfaces did focus either on persons or on cultural objects, but were not able to handle different kinds of entities. The visualizations developed within InTaVia represent objects, persons, but also places and organizations, in a coherent fashion and thereby enable their integrated visual analysis and communication.
(4) While some methods used for the development of theInTaVia are not new, the InTaVia platform is the first of its kind, which supports the full practitioner workflow from searching for information, to the creation of new data, the curation of erroneous data, its visual analysis, and its narrative communication.
(5) The InTaVia project consortium makes its results, methods, and prototypes fully and openly accessible for re-use by the scientific community, European research infrastructures, GLAM institutions, and cultural and creative industries. Open workflows and APIs allow the ingestion of further data to the IKG, the import of local data to the platform, and the re-use in other tools.
With these innovations, the InTaVia consortium expects these developments to have a notable and measurable impact on the local and global promotion of Europe's cultural heritage, on the analysis and understanding of cultural history, on future practices of curating digital assets, and on their use and re-use in various fields including Academia, education, cultural tourism, cultural journalism, and creative industries.