9. CHALLENGE AND OPPORTUNITY
Big Data is growing across Languages, Sectors and Domains
Machine translation,
terminology
annotation, ...
Linked data creation
& processing
GAPS THAT HINDER BUSINESS
Plethora of formats, Adaptability and
platform dependency, Language coverage,
and Usability
10. CHALLENGE AND OPPORTUNITY
Big Data is growing across Languages, Sectors and Domains
Machine translation,
terminology
annotation, ...
Linked data creation
& processing
SET OF GRAPHICAL AND SOFTWARE
INTERFACES
DESIGN DRIVEN BY BUSINESS CASES
14. PUTTING STANDARDS* INTO ACTION
ITS 2.0
Metadata to integrate automated processing
of human language into core Web
technologies
NIF 2.0
Format to achieve interoperability between
NLP tools, language resources and
annotations
OntoLex / Lemon
Model to publish lexica or terminological data
on the web as linked data
Standards deployed in FREME Formats for FREME Enrichment
• HTML5
• General XML
• Selected XML Vocabularies
DocBook, TEI, XLIFF, ODF, …
• Linked Data as turtle, json-ld etc
• More to come – depending on
your needs!
Processing of a growing set of
formats via the Okapi framework
Roundtripping – storing
enrichment in the original content
* Including formal standards and de-facto standards
17. • Machine translation
• Patented technology
• Domain focused translation
• Training data required
Multilingual Enrichment
18. • Harness the power of the still largely untapped
Semantic Web
• Huge amount of data available on the web
• But…largely untapped due to
–Lack of skills
–Lack of awareness
–Lack of tangible use cases
Semantic Enrichment
22. • Example of smart digital content
• Embed semantic features in EPUB3-format
e-book
• Small POC developed using FREME e-services
• E-book on Athens enriched with structured
data from DBPedia
Semantic Book
25. • CKEditor is a WYSIWYG HTML editor widely
used in CMS systems (Drupal, Wordpress)
• We developed a FREME plugin enabling
content authors to use FREME enrichment
services via the GUI
• Demo version available
• Practical application in book creation
CKEditor plugin
27. • Digital book on water sanitation authored by
+160 researchers
• Authoring process involved annotation of
terms using CKEditor with FREME plugin
Use case: Global Water Pathogens Project
30. COGNITIVE CONTENT SOLUTIONS
Cognitive Content: Merge Content Creation and
Content Delivery
Predict What to Publish, Write, Translate next in
order to increase Content related KPIs -> Content
Engagement
Know What Audience Wants What Content
(What/Where/When)
35. THE COGNITIVE STUFF -> Clustering
Dublin | Fine
Art
Dingle| Hiking
Dublin | Pubs
Galway| Oysters
Jameson|
Distillery
Temple Bar|
Music
Guinness|
Visiting Centre
41. Frank Salliau
Project Manager iMinds DSLab
Email frank.salliau@ugent.be
Kevin Koidl
CEO/Founder Wripl Technologies
E-mail: kevin@wripl.com
Twitter: @koidl
Thank you!
Notas del editor
we have used the CKEditor plugin in the case of GWPP. GWPP (Global Water Pathogen Project) is an online resource/book, that is authored by a worldwide community of ~160 researchers on water sanitation. The authoring process happens online with the usage of CKEditor and numerous libraries for content publishing and enrichment. Authors also engage in crowdsourcing of an online glossary of terms that are used in the online book (1.png).
We used the FREME CKEditor plugin to annotate terms of the GWPP glossary within the text of our first published chapters (2.png). The plugin shows up as a simple button on top of the chapter text and provides simple options for users to enrich their material (3.jpg, 4.png). After the process, chapters are annotated with linked terms. Users can hover their mouse over these terms and view their definitions (5.png) or click on them and visit their full description page. The same material is also published in EPUB format (generated by FREME again), with the same hover-behavior on terms (6.png).
Looks at the outside market
Jenny builds these clusters over a few weeks and can start to see trends in the topic clusters which help her predict what user clusters are tending towards what topics.
She can now start studying overlapping topics and see how some topics are more effective and some not.
Looks at the outside market
Jenny builds these clusters over a few weeks and can start to see trends in the topic clusters which help her predict what user clusters are tending towards what topics.
She can now start studying overlapping topics and see how some topics are more effective and some not.
Looks at the outside market
Jenny builds these clusters over a few weeks and can start to see trends in the topic clusters which help her predict what user clusters are tending towards what topics.
She can now start studying overlapping topics and see how some topics are more effective and some not.
Looks at the outside market
Jenny builds these clusters over a few weeks and can start to see trends in the topic clusters which help her predict what user clusters are tending towards what topics.
She can now start studying overlapping topics and see how some topics are more effective and some not.
Looks at the outside market
Jenny builds these clusters over a few weeks and can start to see trends in the topic clusters which help her predict what user clusters are tending towards what topics.
She can now start studying overlapping topics and see how some topics are more effective and some not.