WP5: Media Studies
Work package 5 focuses on the development of the Clariah Media Suite. The Media Suite facilitates access to key Dutch media collections with advanced multimedia search and analysis tools.
The Clariah Media Suite is intended primarily as a research tool for scholars interested in data science with media collections maintained at institutes in the Netherlands such as Netherlands Institute for Sound and Vision, National Library, Eye Film Institute, DANS, Meertens Institute, etc. The Media Suite provides specific tools for researchers to search, select, bookmark, view, annotate, and compare these data. Users with Python skills can work with the Media Suite’s Jupyter Notebooks. All researchers, lecturers and students at Dutch universities can log in to the Media Suite with their university credentials.
Resources in the Media Suite
The Media Suite provides access to, among others:
- The collection of the Netherlands Institute for Sound and Vision with the meta-data of about 1.000.000 radio, television and other AV-items, for a vast majority also with access to streamed content;
- The Jean Desmet film, poster and business archive of Eye;
- Oral history interviews made available through a.o. DANS
- A variety of AV-collections held by other heritage institutes such as soundbites of the Meertens Institute.
The majority of data collections are in the Dutch language.
In addition, the Media Suite provides subtitles, TT888 and ASR-files (automated speech recognition) of AV-items as a search function and as a possible tier for video-annotation in the Media Suite.
Research with the Media Suite
For copyright reasons, the data cannot be exported or downloaded on your own computer. All users have a web-based Workspace within the Media Suite in which they can save their selected items, segmentations and annotations. It is not possible to import your own dataset within the Media Suite.
The Media Suite especially supports ‘close reading’: qualitative content analysis and annotation of audio-visual materials. For exploratory quantitative analysis and ‘distant reading’, we also provide histograms and line charts as visualization and filter options. All Media Suite tools (Search, Explore, Inspect, Compare, Resource Viewer, Workspace) are easy to use and do not require any knowledge of Python. All tools are built with the principle of tool and data criticism in mind; we offer substantial documentation and a metadata inspection tool (Collection Inspector). For more large-scale and advanced quantitative research, users with knowledge of Python can make use of experimental Jupyter Notebooks.
We also offer a platform for datastories within the Media Suite. Researchers can develop stories with data from the Media Suite, for which they can use the Media Suite tools and the Jupyter Notebooks. The datastories enable scholars to share the first results of their research with a wider audience in a multimedia format.
Examples of research using the Media Suite
In previous years, researchers investigated, for instance, the debate on drug regulation in radio broadcasts, news reporting on pandemics on television and WW2 eyewitness testimonies across the media.