Semantic CMS using the Redlink Platform by Thomas Kurz
3 likes898 views
The document discusses semantic content management using semantic web technologies. It outlines how semantic web technologies like linked data, ontologies and natural language processing can help address problems in current content management systems by making content understandable by machines. This allows inferring knowledge from content and identifying relationships between content items. The Redlink platform is presented as integrating Apache Marmotta, Apache Stanbol and Apache Solr to enable semantic lifting, querying and search of linked content. Semantic media asset management is also discussed as a way to bridge the semantic gap for multimedia content.
![Media Fragments
„ … a media-format independent, standard means of addressing
media fragments on the Web using Uniform Resource Identifiers. “
[W3C Recommendation: Media Fragments URI 1.0 (basic)]
!
temporal
t=10,20
spacial
xywh=0,0,20,20
track
track=audio
id
id=chapter2
!
http://www.w3.org/TR/media-frags/
Semantic Content Management
23/36
01/02](https://image.slidesharecdn.com/presentationmodule-140206131713-phpapp01-140227034544-phpapp01/85/Semantic-CMS-using-the-Redlink-Platform-by-Thomas-Kurz-23-320.jpg)
![Media Resource Description
Ontology for Media Resources 1.0
„ … to bridge the different descriptions of media resources, and
provide a core set of descriptive properties.“
[W3C Recommendation: Ontology for Media Resources 1.0]
!
Open Annotation Collaboration
!
!
!
Semantic Content Management
24/36
01/02](https://image.slidesharecdn.com/presentationmodule-140206131713-phpapp01-140227034544-phpapp01/85/Semantic-CMS-using-the-Redlink-Platform-by-Thomas-Kurz-24-320.jpg)
Semantic CMS using the Redlink Platform by Thomas Kurz
- 1. Semantic Content Management Techniques and Tools Talk at MODULE University Vienna Thomas Kurz ! 2014/02/06 Vienna, Austria
- 2. Redlink was founded in 2013/03 and is headquartered in SALZBURG , Austria. John Pereira Aingaran Pillai Andrea Volpini Rupert Westenthaler Jakob Frank Sebastian Schaffert Sergio Fernàndez Semantic Content Management Thomas Kurz David Riccitelli 02/36
- 3. Outline • Why we need Semantics in CMS ? • How can Semantic Web Technologies help ? • How Redlink makes the integration much easier ? ! • Excursus: What about Semantic Media Asset Management Systems Semantic Content Management 03/36
- 4. "We are drowning in information and starved for knowledge." John Naisbitt • Content is highly available through the Internet • Information are distributed over people and systems • Data is available in various media and technical formats We need an efficient way for working with huge amounts of unstructured content Semantic Content Management 04/36
- 5. Content Management Systems • CMS are a single point of entry, providing consistency and the foundations for collaborative work with content • CMS provide functionalities to handle large amounts of content: • • • • Creation of new content Editing of existing content Organisation and management of content Presentation of content • Media-neutral data management (separation of layout and content) Semantic Content Management 05/36
- 7. State of Play in Content Management • Current solutions provide efficient ways to manage content • Domain-specific requirements, like “multichannel content distribution” are addressed • Content can be managed and presented in multi-media formats … B UT … Semantic Content Management 07/36
- 8. Problems in current Content Management Systems • Content is only “understandable” by users and not by machines • • Irrelevant search results Aggregation of relevant content needs to be done manually ! • Inferring Knowledge from Content • Dependencies, relations and inconsistencies among content items need to be identified and defined manually ! • Content is strongly connected to presentation • works only inside a certain environment Semantic Content Management 08/36
- 9. The GOAL It would be right/wrong to sell the product to John Smith. WISDOM John Smith is a potential customer for your products + Insight KNOWLEDGE + Meaning INFORMATION John Smith is a name + Context John Smith DATA Semantic Content Management 09/36
- 10. Slide by Nova Spivack, Radar Networks Semantic Content Management 10/36
- 11. How Semantic Web Technologies can help http://upload.wikimedia.org/wikipedia/commons/f/f3/Semantic_Web_Stack.png Semantic Content Management 11/36
- 12. (Open) Linked Data 1. Use URIs as names for things 2. Use HTTP URIs so that people can look up those names. 3. When someone looks up a URI, provide useful information, using the standards (RDF*, SPARQL) 4. Include links to other URIs. so that they can discover more things. Semantic Content Management 12/36
- 13. Semantic Lifting via Natural Language Processing Semantic Content Management 13/36
- 14. How should we handle this? Semantic Content Management 14/36
- 15. The Redlink Platform Semantic Content Management 15/36 01/02
- 16. The Open Platform for Linked Data http://marmotta.apache.org/ • Read-Write Linked Data • Triple store with transactions, versioning and reasoning • SPARQL and LDPath query languages • Transparent Linked Data Caching Semantic Content Management 16/36
- 17. The Toolbox for Semantic Lifting http://stanbol.apache.org • Semantic Enhancement process chaining • Several Natural Language processing facilities • Multi-language support • Classification and Sentiment Analysis Semantic Content Management 17/36
- 18. The highgly scalable Search Server http://lucene.apache.org/solr/ • Based on Apache Lucene • Many language specific processing procedures • Highly scalable (Solr cloud) and ultra fast • Highly configurable Semantic Content Management 18/36
- 20. PART II ! Media Asset Management Bridging the Semantic Gap
- 21. Semantic Media Asset Management Systems • Multimedia Content is enormously growing within the last decade (Web 2.0) • Multimedia Content must be prepared for automatic processing • for multimedia retrieval • for reuse across platforms, contexts, locations, languages • Multimedia Content Management Systems heavily rely on high quality metadata (meaning is hidden Semantic Gap) Semantic Web Technologies can bridge the gap Semantic Content Management 21/36 01/02
- 22. Where we use Semantics • Controlled Vocabularies • Domain specific Thesauri using standard representations • Reuse of external data • Create Knowledge by linking • (Semi-) Automatic Metadata enrichment and classification • Semantic Search (Facetting, Synonymes, Multilingual) What do we need to bring Media Objects in the Web of Data Semantic Content Management 22/36 01/02
- 23. Media Fragments „ … a media-format independent, standard means of addressing media fragments on the Web using Uniform Resource Identifiers. “ [W3C Recommendation: Media Fragments URI 1.0 (basic)] ! temporal t=10,20 spacial xywh=0,0,20,20 track track=audio id id=chapter2 ! http://www.w3.org/TR/media-frags/ Semantic Content Management 23/36 01/02
- 24. Media Resource Description Ontology for Media Resources 1.0 „ … to bridge the different descriptions of media resources, and provide a core set of descriptive properties.“ [W3C Recommendation: Ontology for Media Resources 1.0] ! Open Annotation Collaboration ! ! ! Semantic Content Management 24/36 01/02
- 26. Hello, my name is Tom! Last summer I was in Paris in France for vacation. It was really amazing. I love Paris! Semantic Content Management 26/36
- 27. RDFize Tom's statement Tom likes Paris, France. -‐> Tom likes Paris. -‐> ( Tom, likes, Paris ) -‐> Paris is a part of France. -‐> ( Paris, partOf, France ) Semantic Content Management 27/36
- 28. RDFize Tom's statement Tom likes Paris, France. -‐> Tom likes Paris. -‐> ( Tom, likes, Paris ) -‐> Paris is a part of France. -‐> ( Paris, partOf, France ) Semantic Content Management 28/36
- 29. Link to external resources Semantic Content Management 29/36
- 30. But what about this? Title:! „Me and the big thing“! Album:! „A vacation in Paris“! Author:! „Tom Tester“! Semantic Content Management 30/36
- 31. Extract Information Title:! „Me and the big thing“! Album:! „A vacation in Paris“! Author:! „Tom Tester“! Semantic Content Management 31/36
- 32. Link Information :image! :hasFragment! :image#xywh=...! ! :image#xywh=...! :subject foaf:Person! ! :image ! :hasFragment ! :image#xywh=..! ! :image#xywh=.. ! :subject ! dbpedia:EiffelTower! (50%) Title:! „Me and the big thing“! Album:! vacation in <http://dbpedia.org/resource/Paris>“! „A Author:! „<http://tom-tester.org/me>“ Semantic Content Management 32/36
- 34. Create new facts by using Contextual Semantics :image :hasFragment :image#xywh=...! :image#xywh=... :subject foaf:Person! :image :author tom:me! ! :image :hasFragment :image#xywh=..! ! :image :location dbpedia:Paris! :image :location geonames:France! ! :image#xywh=.. :subject dbpedia:EiffelTower! (+90%)! ! :image :showsOnTheLeft tom:me (50%)! :image :showsOnTheRight dbpedia:EiffelTower! :image :showsOnTheRight dbpedia:VisitorAtraction! Semantic Content Management 35/36
- 36. Thanks for your attention! Any Questions? thomas.kurz@redlink.co Semantic Content Management