Die Crossmediale Suche der ARD

ARD – The Media Data Hub is a key project for the ARD structural reform.

Started as a cross-media metadata system, the Data Hub is the basis for modern and connected work. This is associated with significant increases in efficiency in the storage and retrieval of content. The automatic generation of metadata (e.g. keywording) and the provision of intelligent, cross-media search functions provide authors with convenient access to ARD’s entire program treasure trove. Increasing automation in the areas of editing, documentation and archives as well as IT lead to a significant reduction in effort. The MDH was initially developed on behalf of WDR and in collaboration with the Fraunhofer Institute for Intelligent Analysis and Information Systems IAIS.

ARD faces the challenge of a growing number of internal systems, information and distribution channels. In addition to the design of the content, the effort required for documentation and information obligations is also increasing. Rights clearance, copyright disclosure, Gema notifications or settlements are essential parts of the workflow. In order to produce high-quality content and to cope with the increased documentation requirements, ARD editors need a variety of special systems. Researching information in these systems is costly and time-consuming. Time that is missing for creative work processes. Along the media production chain, the Media Data Hub, as a central data hub, facilitates access to internal systems such as licensing, planning, production and billing systems, and information from external data sources such as the web.

The integration of data from the aforementioned special systems enriched by automatically generated metadata based on AI processes enables editorial teams to conduct cross-system, efficient and cross-media research, viewing, playout and editing. Cross-domain collaboration is greatly simplified, and authors are given access to the entire program repository. Through semantic search methods and AI-supported evaluation of user actions, the system enables more precise search results and additionally facilitates the research of content through user-adapted search suggestions.