ºÝºÝߣ

ºÝºÝߣShare a Scribd company logo
Multilingual Value Chain Solution for
the Digital Single Market
Federated Active Linguistic data CuratiON
EU FP7 Project
The FALCON project combines the power of open data on the web
with data-driven language technologies to construct the Localization
Web.
Partners:
Wholesalers
Digital Single Market
Decoupage.ie
Wholesalers
ecommerce
SaaS
ePayment
Service
Customer
? Digital Single Market works well downstream
¨C English as lingua franca
¨C Export-focused Medium-SME wholesalers and service providers
¨C MNCs with establish multilingual offering
? Language Barrier firmly in place upstream
? Challenge to Customer Engagement Ecosystem
¨C Must become systematically multilingual
¨C Must serve micro-domains that allow SMEs to add value
Wholesalers
Wholesalers
Customer Engagement Ecosystem
Niche Value Add
Social
Media
Online
Communities
Search
& SEO
Content
Analytics
Trade Guilds/
Associations
Events
Knowledge &
training resources
Translation
Translation workflow
The company has also reduced its production
capacity by ceasing manufacture of chest
freezers and freestanding microwave ovens
Extraction &
Segmentation
production capacity
capacit¨¦ de production
?
? Annotation with
Existing Terms
chest freezer
microwave oven
r¨¦frig¨¦rateur
four ¨¤ micro-onde
?
?
?
?
Auto suggestion from
Babelfy/Babelnet
D'autre part, la soci¨¦t¨¦ a r¨¦duit sa capacit¨¦ de
production en arr¨ºtant la production de
r¨¦frig¨¦rateur et de fours micro-onde pose-libre
Machine Translate
with Term Translations
MT Vendor?
D'autre part, la soci¨¦t¨¦ a r¨¦duit sa capacit¨¦ de
production en arr¨ºtant la production de
cong¨¦lateurs coffres et de fours micro-ondes
pose-libre
?
cong¨¦lateurs coffres
fours micro-ondes
?
Postedit and capture
terms in context
??
?
?
?
?
PE
PE
PE
PE
PE
PE
PE
?
PE
?
? Protect & pool niche
knowledge
? Interlink corpora and
lexical-conceptual
resources
? Measuring ROI at each
point in value chain
? Manage ownership,
rights and rewards
? Privacy by Design for
Social Media data
resources
? Open data for NLP
shared task
Integrated Content/Data Value Chain
Public Data
Content publisher
Support Service
Provider
Language
Technology
Provider
? Better in-context postediting:
¨C XTM-Easyling
? Feeding term suggestions from posteditor to Terminology
Management
¨C XTM-Interverbum
? Dynamic Retraining
¨C XTM-DCU
? Bilingual Dictionary SMT improvements
¨C XTM-DCU
? NER, terminology enforcements, forced decoding
¨C XTM-Interverbum-DCU
? Postediting prioritisation and term flagging
¨C TCD-DCU-XTM
? Publishing interlinks of parallel text, lexically rich term bases
¨C TCD: DG-T TM, EurVoc, Snomed-CT, LEMON, BabelNet
FALCON Innovation
Terminology Management
Website in context translation
? THANK YOU!

More Related Content

Falcon

  • 1. Multilingual Value Chain Solution for the Digital Single Market Federated Active Linguistic data CuratiON
  • 2. EU FP7 Project The FALCON project combines the power of open data on the web with data-driven language technologies to construct the Localization Web. Partners:
  • 3. Wholesalers Digital Single Market Decoupage.ie Wholesalers ecommerce SaaS ePayment Service Customer ? Digital Single Market works well downstream ¨C English as lingua franca ¨C Export-focused Medium-SME wholesalers and service providers ¨C MNCs with establish multilingual offering ? Language Barrier firmly in place upstream ? Challenge to Customer Engagement Ecosystem ¨C Must become systematically multilingual ¨C Must serve micro-domains that allow SMEs to add value Wholesalers Wholesalers Customer Engagement Ecosystem Niche Value Add Social Media Online Communities Search & SEO Content Analytics Trade Guilds/ Associations Events Knowledge & training resources Translation
  • 4. Translation workflow The company has also reduced its production capacity by ceasing manufacture of chest freezers and freestanding microwave ovens Extraction & Segmentation production capacity capacit¨¦ de production ? ? Annotation with Existing Terms chest freezer microwave oven r¨¦frig¨¦rateur four ¨¤ micro-onde ? ? ? ? Auto suggestion from Babelfy/Babelnet D'autre part, la soci¨¦t¨¦ a r¨¦duit sa capacit¨¦ de production en arr¨ºtant la production de r¨¦frig¨¦rateur et de fours micro-onde pose-libre Machine Translate with Term Translations MT Vendor? D'autre part, la soci¨¦t¨¦ a r¨¦duit sa capacit¨¦ de production en arr¨ºtant la production de cong¨¦lateurs coffres et de fours micro-ondes pose-libre ? cong¨¦lateurs coffres fours micro-ondes ? Postedit and capture terms in context ?? ? ? ? ? PE PE PE PE PE PE PE ? PE ?
  • 5. ? Protect & pool niche knowledge ? Interlink corpora and lexical-conceptual resources ? Measuring ROI at each point in value chain ? Manage ownership, rights and rewards ? Privacy by Design for Social Media data resources ? Open data for NLP shared task Integrated Content/Data Value Chain Public Data Content publisher Support Service Provider Language Technology Provider
  • 6. ? Better in-context postediting: ¨C XTM-Easyling ? Feeding term suggestions from posteditor to Terminology Management ¨C XTM-Interverbum ? Dynamic Retraining ¨C XTM-DCU ? Bilingual Dictionary SMT improvements ¨C XTM-DCU ? NER, terminology enforcements, forced decoding ¨C XTM-Interverbum-DCU ? Postediting prioritisation and term flagging ¨C TCD-DCU-XTM ? Publishing interlinks of parallel text, lexically rich term bases ¨C TCD: DG-T TM, EurVoc, Snomed-CT, LEMON, BabelNet FALCON Innovation
  • 8. Website in context translation