|Sector||All areas of publishing|
|Target Users||Digital content managers, software developers/integrators, ICT Staff at publishing houses|
The FREME Horizon 2020 innovation action aims at enabling publishers to make their digital content ready for a multilingual market by offering easy access to machine translation services. It also aims at giving publishers tools to enrich their content with data from the enormous vault of high-quality online datasets.
FREME is a multidisciplinary and transdisciplinary action and the FREME consortium brings together knowledgeable specialists in language resources, tools, and services, multilingualism, infrastructure, standardisation, semantic web, linked data and open data including public sector information, publishing, localisation, and business development.
The general objective of FREME is to build an open innovative commercial-grade framework of web-based e-services for multilingual and semantic enrichment of digital content.
The FREME consortium consists of the following partners: DFKI (Germany), Agro-Know (Greece), iMinds (Belgium), Infai (Germany), ISMB (Italy), Tilde (Latvia), VistaTEC (Ireland), wripl (Ireland).
The rise of the digital industry poses a major disruption for the publishing industry. As consumers are steadily shifting towards digital consumption of published content, revenues from traditional printed media are in decline. The publishing industry suffers from a price paradox: consumers are less prepared to pay for digital content such as e-books, online news articles, but production costs are still high.
In order to decrease production costs, publishers are pressured into making their content production workflows more efficient, e.g. by moving away from a print-first workflow and adopting a multi-channel workflow.
Increasing competition forces publishers to increase their sales volumes. This creates a need for:
- improved discovery of their content;
- providing customers with richer and more valuable content;
- new reading experiences;
- interoperable formats.
Furthermore, publishers are pressed to enter new and multilingual markets, increasing the need for effective translation and localization of their content.
FREME provides an opportunity for publishers to improve their content and workflows by integrating semantic and multilingual services into their content production software. The FREME platform will enable publishers to enrich their content by integrating FREME e-services in existing publishing workflows via lightweight interfaces and open content formats.
The FREME framework will integrate the following e-services based on existing and mature technologies:
- e-Link: enrich content with information from an external data source;
- e-Entity: recognize and disambiguate named entities (people, places etc...);
- e-Terminology: terminology management and annotation;
- e-Translation: cloud based machine translation service;
- e-Publishing: publishing digital content in EPUB 3 format.
The FREME concept and approach are designed from the perspective of real world business needs. Multilingual and semantic technology providers will be able to integrate their enrichment services into the FREME framework within any data value chain in a user-friendly and robust way.
The FREME framework is being validated in 4 real world business cases:
- Publishing and content creation;
- Content translation/localisation;
- Cross-language content sharing and access;
- Personalised Web content recommendations.
The role of technology
Although the FREME framework for multilingual and semantic enrichment of digital content integrates e-services based on existing and mature technologies, the innovative character lies in the way e-services are integrated using standard formats and well-described REST API's.
The framework will integrate the following e-services:
- e-Internationalisation based on the technologyInternationalisation Tag Set (ITS) Version 2.0;
- e-Link based on theNatural Language Processing Interchange Format (NIF), the DBpedia Ontology and RDF;
- e-Entity based on entity recognition software and existing linked entity datasets fromLinked Data and Open Data;
- e-Terminology based on cloud terminology services for terminology management and terminology annotation web service to annotate terminology in ITS 2.0 enriched content;
- e-Translation based on cloud machine translation services for building custom machine translation systems;
- e-Publishing based on cloud content authoring environment (for example e-books, technical documentation, marketing materials etc.) and its export for publishing in Electronic Publication (EPUB3) format.
At the time of writing, FREME is approaching the end of its first year.
Results relevant for publishing are:
- An alpha version of the FREME framework has been developed;
- The framework has been integrated in a pilot content authoring environment developed at iMinds.
Development of an open source plugin for a widely used web content editor is underway.