Freme at feisgiltt 2015 freme use cases

FREME Business Cases – FREME at FEISGILTT 2015 WWW.FREME-PROJECT.EU 1

Co-funded by the Horizon 2020Framework Programme of the European UnionGrant Agreement Number 644771

FEISGILTT 2015 |BERLIN, 2 JUNE 2015

Felix Sasaki, DFKI / W3C FellowOn behalf of the FREME Consortium

FREME BUSINESS CASES

www.freme-project.eu

http://www.freme-project.eu/


THE FREME PROJECT

• Two year H2020 Innovation action; start February 2020

• Industry partners leading four business cases arounddigital content and (linked) data

• Technology development bridging language and data

• Outreach and business modelling demonstrating monetization of the multilingual data value chain


GROUPING OF PARTNERS

• Technology solution providers◦ DFKI, InfAI, iMinds, Tilde

• Business Case Partners◦ iMinds, VistaTEC, Agro-Know, Wripl

• Dissemination and Business Modeling◦ ISMB


CHALLENGE AND OPPORTUNITY: BIG DATA IS GROWING ACROSS LANGUAGES, SECTORS AND DOMAINS

• BC: Digital publishing

• BC: Translation and localisation

• BC: Agriculture and food domain data

• BC: Web site personalisation

Agriculture metadata, user content, news

content, …

WHAT LIES AHEAD FOR SEVERAL INDUSTRIES? SEE THE FREME BUSINESS CASES

FR

ES JA, ZH, ...

AR


CURRENT STATE OF SOLUTIONS

Machine translation, terminology

annotation, ...

Linked data creation & processing

GAPS THAT HINDER BUSINESS:

• Plethora of formats• Adaptability and platform dependency• Language coverage• Usability “The right tool for the right person

in given and new enterprises”: technology influences job profiles


FREME TO THE RESCUE: ENRICHING DIGITAL CONTENT

Machine translation, terminology

annotation, ...

Linked data creation & processing

LT and LD as first class citizens on the Web

A SET OF INTERFACES* - DESIGN DRIVENBY BUSINESS CASES

LT and LD for varioususer types: (application) developer, content architect, content author, translator, …

* Graphical interfaces* Software Interfaces



FREME E-SERVICES – BIRDS EYE VIEW

• e-Entity◦ Automatic annotation of named entities

• e-Terminology◦ Annotation of terms and linkage to term databases

• e-Link◦ Enrichment with information from (linked) (open) data sources*

• e-Translation◦ Cloud based machine translation

• e-Internationalisation◦ ITS 2.0 metadata to govern the multilingual & semantic content workflow

• e-Publishing◦ Publish enriched content in ePub format

* More on this: see presentation FEISGILTT 4 June


BUSINESS CASES


AUTHORING AND PUBLISHING MULTILINGUALLY AND SEMANTICALLY ENRICHED EBOOKS

• Challenge◦ Moving to digital publishing workflows leads to high costs. Added value beyond the content

itself is needed.

• Solution◦ Easy to integrate solutions, based on standardised interfaces for software components and

authors, responding to usability requirements


AUTHORING AND PUBLISHING MULTILINGUALLY AND SEMANTICALLY ENRICHED EBOOKS

• Example: Integration into ePub editing mode of oXygen XML Editor

e-Entity: annotate named entities


INTEGRATING SEMANTIC ENRICHMENT INTO MULTILINGUAL CONTENT IN TRANSLATION AND LOCALISATION

• Challenge◦ Localisation service providers are under pressure: more speed, better quality, lower

payment rates per translation outcomes. LSPs need to offer additional services, going beyond translation

• Solution◦ Allow translators to work with linked data sources during the translation process, to add

additional information to the translated content.



• Example: Integration into XLIFF 2.0 editing mode of oXygen XML Editor

• Combination of services◦ e-Entity: annotate named entities◦ e-Link: fetch additional information from a linked data source like dbpedia, specific to the

type of entities (places, persons, …)



• Enriching content with machine readable information – represented as JSON ◦ Input: “Welcome to Berlin … Marlene Dietrich!”◦ Output:

[ { "@id": "dbpedia:Marlene_Dietrich", "@type": "person", "born": "1901-12-27" }, { "@id": "dbpedia:Berlin", "@type": "place", "population": "3746444" }

]

May be basis e.g. formultilingual generation:• “… born 1901”• “… geboren 1901”• “…１９０１年生まれ”• …


ENHANCING THE CROSS LANGUAGE SHARING AND ACCESS TO OPEN AGRICULTURAL AND FOOD DATA

• Challenge◦ Content metadata is not available in multiple languages, needs to be curated manually, and

the metadata is not linked to external data sources.

• Solution◦ Via FREME, make metadata available in several languages◦ Easy data discovery for users with different linguistic profiles◦ Allow to reach out to new markets


FREME-EMPOWERED PERSONALISED CONTENT RECOMMENDATIONS

• Challenge◦ Personalised content recommendations for content rich websites help to increase

engagement of users on a website. Current solutions focuses on English web sites.

• Solution◦ Using FREME, content recommendation will be provided in several languages◦ The e-Services e-Entity and e-Translation will help to identify user relevant items in Web

content


WANT TO TRY THINGS OUT?

• Go to http://api.freme-project.eu/doc/0.1/

• Check out API demo calls

• More tomorrow = FEISGILTT day 2, during the “linked data” presentation

http://api.freme-project.eu/doc/0.1/

http://api.freme-project.eu/doc/0.1/


DEMO:ENRICHMENT IN OXYGEN XML EDITOR

DISCLAIMER: NOT YET FREME BASED, BUT SHOWS THE POSSIBILITIES OF COMBININGE-ENTITY AND E-LINK


LESSONS TO LEARN FOR PROJECTS INVOLVING ENRICHMENT

• Adequate interfaces for the user type in question are key◦ E.g. content creator vs. translator

• Hide complexity of technologies in a user specific way◦ Ability to adapt linked data queries without knowing linked data – “I want to find

information about cities and related tourists sites in the radius of X miles”

• New job roles may be needed◦ Today: content architect, set up of content workflows◦ With FREME: enrichment architect, set up of language and data tooling (e.g. preparing

linked data queries)

• FREME may also help to automatize the collaboration between departments (e.g. localisation <> marketing) within one organisation


CHALLENGES

• Technical, e.g.: How to store enrichment information in the format being enriched?◦ RDF as storage and processing format not widely deployed e.g. in multilingual content

creation

• Closing knowledge gaps◦ Sharing knowledge between different types of tool developers: language technology, linked

data, localisation, content creation, …

• Aspects of data sets◦ Availability for commercial purposes, quality, sustainability, complexity of the data (“How

do I find X?”)

• Qualification & tooling◦ Teaching translators just enough linked data to do proper enrichment

• Convincing (end) customers: developing killer applications using FREME


NEXT STEPS FOR FREME

Creating & updating prototypes of e-Services

Agile approach: gathering feedback from business cases feeds implementations

Potential input to standardisation of the open Web platform: ease access to data (sources + technologies) and language technologies for various user types

Learn more at www.freme-project.eu

Idea: discuss concrete standardisation opportunities at a W3C workshop on interfaces to LT & LD technologies? Ping me if interested


CONTACTS

FELIX SASAKI

Senior Researcher DFKI / W3C Fellow

On behalf of the FREME consortium

E-mail: [email protected]

CONSORTIUM

http://www.tilde.com/

http://www.iminds.be/

http://www.agroknow.gr/agroknow

http://wripl.com/

http://www.vistatec.com/index.html

http://infai.org/de/Aktuelles

http://www.ismb.it/

http://www.dfki.de/web

Internet

Freme at feisgiltt 2015 freme use cases