MEDIA DATA HUB

Automatisierte, digitale Prozesse
für die Medienbranche.

Cross-media research and automated, digital workflows for the editorial workplace.

Our Success Story

Discover, arrange and select content

Consistent Cross-media Viewpoint and Central Access Point

A metadata hub between existing IT systems and innovative applications.

AI Assistance

AI assistants automatically generate metadata.

Semantic Cross-Media Search Capability

AI based research for cross-media portfolios

A Data Hub and Digital Assistant for Broadcasters and Publishers

A particular challenge for media companies is being able to provide and access high-quality metadata conveniently. This is a challenge that is constantly growing with the number of sources of information and distribution channels available today.

Access to this metadata is an essential part of all internal processes such as planning, production and distribution. There are also some important accompanying processes such as research, rights clearance and billing. With the Media Data Hub, metadata can be collected, generated and managed throughout the company.

It combines relevant information across the various specialized platforms and thus provides a uniform, cross-media view of media topics. Missing metadata is automatically supplemented using data analysis, artificial intelligence and linked data methods. As a hub for metadata, the Media Data Hub serves both as a data provider for third-party systems and as a central platform for data mining services, algorithms and models.

Simplify Editorial Workflows

The system consists of more than 40 different microservices and micro frontends, as well as cluster computing applications designed to intelligently process large amounts of data.

One of the key platform features is the integration of varied data sources from legacy systems, including web and social media. All sources are displayed on a graph-based data model that enables users, systems and also algorithms to access the harmonized data in a consistent and homogeneous way. This is a key prerequisite for the use of modern analysis algorithms that enrich and link data objects across different sources. The Media Data Hub has connectors to link mainstream software systems and internet data sources.

A large number of analysis modules can be seamlessly integrated into a data import. Modules for keyword extraction, semantic concept extraction, Wikipedia disambiguation, duplicate detection and similarity calculation are just a few of the many additional options.

The Media Data Hub offers an information retrieval interface. This provides a comprehensive API (application programming interface) suitable for any web or mobile-based search application. In addition to the standard options, advanced features such as similarity search, hit aggregation, geospatial queries and many more can be configured. The cornerstone of the retrieval component is a training workflow to customize the search result ranking into defined target groups based on artificial intelligence.

In summary: The Media Data Hub is the perfect choice for broadcasters and publishers as a starting point for efficient automation, data and service strategies. The Data Hub can be integrated into any existing system landscape. We fully customize each hub according to our customer’s needs and develop it as required.

The structure supports key use cases:

  • Cross-media research in media archives
  • Data collection for accounting workflows (e.g., Gema)
  • The attainment of information claims (e.g., copyright information)
  • The generation of metadata for video SEO
FAQ

Frequently Asked Questions.

How can I ensure that unauthorized persons cannot access the data?

Authentication and authorization take place in the Media Data Hub via the OpenConnect and OpenID protocols. The services are connected to the company’s own user directory and can map any authorization structures. Data is released using the global OAUTH2 standard.

How is data synced between legacy systems and the Media Data Hub?

This is scheduled using the PULL method, event driven, or using the PUSH method using the REST-based import interface.

Which data types and data formats are supported by the Media Data Hub?

The Media Data Hub is designed as a metadata hub, but in principle also supports unstructured data of all media types. Connectors and transformers are already available for a variety of data types and formats (XML, JSON, etc.), which can easily be adapted to specific sources.

Which data analytics services can complement my Media Data Hub?

The Media Data Hub has a variety of microservices for text, audio, image or video analytics. In addition, higher-order services can be connected to analyse user feedback for task-specific learning.

Are you interested? Let's talk.