Dr. David Lewis

Summary

The ADAPT Centre will address the grand challenge in Digital Platform Content and Applications: how can enterprises, institutions and individuals easily assimilate, reuse and interact with the global torrent of digital content? Transitioning from the Centre for Global Intelligent Content(CNGL), ADAPT will build upon demonstrated, world-recognised research and significant commercial impacts of the multidisciplinary applicant group to develop breakthroughs in Global Digital Content technology, which will turn this heterogeneous torrent of content into meaningful global conversations between organisations and users. This will be achieved by making unprecedented advances in the areas of multilingual natural language processing (NLP), multimedia content analysis and transformation, personalisation and multimodal delivery. The ADAPT Centre's research programme will enable: deeper understanding of multilingual content through multilingual NLP; dynamic transformation of multimedia content to break down language and cultural barriers; personalisation of the user experience to ensure rapid assimilation and reuse of content; and multimodal/multimedia interaction with global content for natural user engagement. ADAPT will deliver measurable world-class research and economic impact: 1650 publications, 200 high-quality postgraduates, 12 spin-out companies, 80 licences to industry and over 500 jobs. This will revolutionise Global Digital Content R&D with unprecedented societal impact and benefit to the Irish economy, and beyond. The ADAPT Centre helps industry and individuals understand, manage and control the overwhelming amount of content with which we are all increasingly confronted. It revolutionises conventional models of content production, analysis and delivery by embedding rich knowledge, automatic understanding, translation, personalisation and delivery capabilities in Global content. ADAPT empowers enterprises, institutions and individuals to seamlessly create, modify and share content in the way most suited to their preferences and needs. It unites world-leading experts to concretely address important research and industry needs identified by the Irish Government. ADAPT will deliver cutting-edge R&D with benefits to Irish industries, institutions and individuals.

Funding Agency

Science Foundation Ireland

Programme

SFI Research Centre

Project Type

SFI Research Centre

Summary

SFI CSET 2nd Plase funding: One of the most significant changes to people's lives in recent years has been the explosion of content available to users, enterprises and communities. The Content Revolution is at the heart of a new, empowering trend which sees enterprises and users in new roles as creators, curators and consumers of content, in social and corporate contexts. The scale of this change is astonishing: over 175 million tweets will be sent per day in 2012 (Solis, 2012), with the majority in languages other than English. Today Facebook alone is larger than the entire World Wide Web (WWW) in 2004 (Solis, 2011). New content is not restricted to text, but includes richer media, with 60 hours of video uploaded every minute, or one hour every second to YouTube. (YouTube, 2012). Filtering and personalised retrieval are emerging as key trends in accessing content, with Google using as many as 57 different signals to influence search (Pariser, 2011)(Boyd, 2011). However, these trends do not address the desire of organisations, communities and individuals for content and services to be delivered in their own language, according to their own needs, preferences and context. Fundamental challenges arise in the way content can be dynamically created, curated, processed and delivered for global usage. CNGL II (Centre for Next Generation Localisation II) seeks to revolutionise the way people can seamlessly interact with content, systems and each other to achieve unprecedented levels of access, efficiency and empowerment. Breakthroughs in the processing, unification and seamless integration of multilingual, multi-modal and multimedia content are key goals in enabling the global content industry and society of the future. CNGL II will pioneer a new concept called 'Global Intelligent Content'. Global Intelligent Content will enable digital content to be more discoverable, adaptable and repurposeable. By embedding new levels of knowledge and intelligence into the content, it will enable advanced intelligent content services to automatically process and transform that content. Global Intelligent Content will thus enable itself to be actively discovered and transformed by advanced intelligent content services so as to make that content more discoverable, semantically rich, adaptable, contextually aware and reusable across global markets. This represents a major shift in the way content is managed, creating a more robust, agile and interoperable relationship between content and the value chain of services acting upon it. In this way Global Intelligent Content can be dynamically transformed based on current user interaction, perceived user intention or current delivery context.

Funding Agency

SFI

Programme

CSET

Summary

Language barriers constitute a formidable obstacle to the free flow of information, products and services in an increasingly globalised economy and information society. "Localisation" refers to the process of adapting digital content to culture, locale and linguistic environments at high quality and speed. Localisation is a key enabling, value-adding, multiplier component of the global software and content distribution industry. Localisation seeks to overcome language barriers. The Centre for Next Generation Localisation (CNGL) is a Academia-Industry partnership with over 100 researchers developing novel technologies addressing the key localisation challenges of volume, access and personalisation. Its objective is to produce substantial advances in the basic and applied research underpinning the design, implementation and evaluation of the blueprints for the Next Generation Localisation Factory. Our mission is to revolutionise localisation via breakthroughs in automation, composition and integration, focusing on: (i)Integrated machine translation technology, (ii) Speech-based interfaces and more personalised speech output, (iii) Multilingual digital content management for personalised multilingual content access and delivery, (iv) Localisation workflows and system integration. Dr Lewis' team focusses on the latter, systems integration, including the models and methods for the rapid integration of workflows, service oriented architectures and service management.

Funding Agency

Science Foundation Ireland

Programme

Centres for Science, Engineering and Technology

Summary

Federated Active Linguistic data CuratiON: This project will deliver a modular software platform that will enable SMEs in the localisation industry to leverage open language resources published by public bodies. This platform will bootstrap a commercially sustainable localisation process that allows SME Language Service Providers (LSPs) to continuously reuse the output of web content translation projects. By integrating these outputs with language resources from public bodies, SME LSPs can easily and cheaply optimise and tailor open source machine translation and automated terminology extraction components to the needs of their customers. Today, the curation of parallel text involves the publication of the undifferentiated results of translation projects in differing formats or via centralised repositories. This restricts SMEs in leveraging these resources to train machine translation engines to match incoming translation projects. FALCON will enable the controlled sharing and reuse of language resources, combining open corpora from public bodies with richly annotated output from commercial translation projects. Federated access control will enable sharing and reuse of commercial resources while respecting business partnerships, client relationships, competitive and licensing concerns. Fine grained traceability language resource reuse will enable return on investment calculations on resource curation in public bodies and SME LSPs alike. The consortium combines academic expertise in localisation interoperability, linked data management, statistical machine translation and statistical parsing with SMEs offering services and solutions for translation and terminology management. This project will enable the language service industry to realise its potential as a major sources of shared linked data by measurably improving cost, speed and quality benefits of reuse. The consortium members will then be strongly placed to commercially exploit these benefits.

Funding Agency

EU

Programme

FP7-ICT-2013-SME-DCA

Project Type

STREP

Summary

Linked Data as an enabler of cross-media and multilingual content analytics for enterprises across Europe: The explosive growth of content in volume, velocity and variety on the Web demands new approaches to content analytics, addressing issues in large scale analysis and interpretation of heterogeneous data sets, originating in different media, human languages, jurisdictions, etc. Among these, language diversity in particular has become a ubiquitous aspect of the Web in light of increasing globalization. Recently, semantic-level, i.e., language- and media-independent, data analysis and representation methods such as those provided by Linked Data and Semantic Web technologies, have been introduced to provide innovative content analytics solutions for such heterogeneous, multilingual and multimedia content. An important missing component however, is the representation of language- and media-specific information that will be needed for interpreting such data correctly - across different media and across the increasing variety of human languages used nowadays on the Web. In this proposal we therefore aim at addressing this missing component by providing an ecosystem for the establishment of a new Linked Open Data (LOD) based ecosystem of free, interlinked, and semantically interoperable language (corpora, dictionaries, lexical and syntactic metadata, etc.) and media (image, video, etc. metadata) resources that will allow for free and open exploitation of such resources in multilingual, cross-media content analytics across the EU and beyond, with specific use cases in different industries. LIDER addresses three key objectives: 1. To support multilingual and multimedia content analytics in enterprises. 2. To provide a discussion platform for the development of novel LOD based multilingual, cross-media content analytics systems that will be able to use Free, Open and Interoperable language and media resources and services. 3. To reach consensus on short and long term goals in LOD-based multilingual, cross-media content analytics in enterprises.

Funding Agency

EU

Programme

FP7-ICT-2013-10

Project Type

Coordination and Support Action