IcemanMediaHub™ is an event driven solution to manage content as well as a developer driven toolbox to create and manage services and applications using content. IcemanMediaHub™ does not store any content, but provide a library of services containing content. IcemanMediaHub™ will exchange and provide content from a variety of services to a large number of content users (applications) based on subscriptions. Since IcemanMediaHub™ is event driven it will deliver content via push and in real time, to subscribing applications or services, whenever it is triggered by an event.
The role of a registered service is to provide content in a desired format. A service will deliver content to IcemanMediaHub™ (ingestion of text,images, video, triggered actions from social media etc) or provides desired functionality (such as ”search”, ”store”, ”carry out a certain action”). A service can also receive/handle content provided by IcemanMediaHub™ through subscriptions from other services.
An application is a user of services. A registered application is subscribing for events in one or several services. A service is providing content to IcemanMediaHub™, while an application is a subscriber and user of content.
One of the main obstacles in all editorial environments is the lack of standards regarding the format of content. The majority of all CMS solutions provide different XML standards, which makes sharing of content between different CMS solutions very difficult. IcemanMediaHub™ has an inbuilt XML exchange service that turns any XML format into NewsML, a well defined and de facto standard format. With this service all CMS solutions can share content and communicate with each other through IcemanMediaHub™ without beeing integrated to each other. This will enable the editorial environment to use different CMS solutions, add new solutions and to dismiss others, without timeconsuming and expensive integrations. (Read more about NewsML)
Services can be configured to use text analysis for text categorization, text clustering, concept/entity extraction, production of granular taxonomies, sentiment analysis and document summarization. Currently a integration with Magnet from Klangoo is available.
A schematic overview over the text analysis process is found here.