16

VRT: Use Case Textual data: a lot of it VRT creates very large amounts news data (textual) –Mostly News or related –sporza.be –deredactie.be –cobra.be

Embed Size (px)

Citation preview

VRT:Use Case

Textual data: a lot of it

• VRT creates very large amounts news data (textual)– Mostly News or related– sporza.be– deredactie.be– cobra.be– …

Data management

Challenge: automated tagging & categorization

• Tagging and categorizing the content is a resource intensive task (manpower)!

• Tagging consistently with a team is a difficicult challenge!

• How to handle the existing archive?

Test case

• Content from:– sporza.be– deredactie.be– cobra.be

• Over 100.000 articles

• Automated Named Entitiy recognition• Automated Categorization

• Available in search-interface

Named Entity Recognition

Categorization or Topic Selection

Why?

Get a better understanding of the metadata management of an organization like VRT

Showcase our solution

Explore opportunities with VRT & other partners

Benefits &Applications

Benefits & Applications

• Uniform annotations over all content• Processing of backlog• Time savings:

– 1 min article– 100.000 articles– 208 working days

• Personalized filtering• Automated publishing• Easy search• Simplify content reuse

Zeticon

MediaHaven MAM

• Media Asset Management

• Problematic content reuse• Inefficient use of time• Security & Rights Management• Scalability

MediaHaven Analytics

• Information retrieval

• Resource Intensive• Consistency• Scalability• Language dependencies

www.mediahaven.com