- Date:
- 2006-2009
- Themes:
- Machine Translation, Machine Learning
- Funding:
- EU
More than half of the EU citizens are not able to hold a conversation in a language other than their mother tongue, let alone to conduct a negotiation, or interpret a law. In a time of wide availability of communication technologies, language barriers are a serious bottleneck to European integration and to economic and cultural exchanges in general. More effective tools to overcome such barriers, in the form of software for machine translation and other cross-lingual textual information access tasks, are in strong demand.
Statistical methods are promising, in that they achieve performances equivalent or superior to those of rule-based systems, at a fraction of the development effort. There are, however, some identified shortcomings in these methods, preventing their broad diffusion. As an example, even though lexical choice is usually more accurate with Statistical Machine Translation (SMT) systems than with their rule-based counterparts, the text they produce tends to be less fluent. As a second example, SMT systems are trained in batch mode and do not adapt by taking user feedback into account. Finally, in Cross-Language Information Retrieval tasks, query words are most often translated independent of one another, thus giving up possibly relevant contextual clues.
SMART is an attempt to address these and other shortcomings by the methods of modern Statistical Learning. The scientific focus is on developing new and more effective statistical approaches while ensuring that existing know-how is duly taken into account. By bringing together leading research institutions in Statistical Learning, Machine Translation and Textual Information Access, the SMART consortium is well positioned to achieve this goal.
Thorough field evaluation on three user scenarios, involving user groups from innovation-oriented SMEs, and extensive exploitation and dissemination activities will ensure that advances make their way out of the laboratories, in the form of both significant and measurable improvements over existing technologies and of new applications currently beyond the state of the art.
- A first user scenario focuses on the work of professional translators and aims at validating new technologies by assessing impact on productivity.
- The second scenario considers the work of technicians providing support to customers over the phone, holding a conversation in a language different from the language the technical documentation available to him/her is written in.
- The third user scenario, finally, consists in enabling a user to access portions of the multilingual Wikipedia in languages of which (s)he has limited command.
SMART is a 3-year "Specific Target Research Project" (STReP) funded by the European Commission through its "Information Society Technologies" (IST) priority, as part of the sixth Framework Programme. It started on October 1, 2006 and is coordinated by Nicola Cancedda at Xerox Research Centre Europe.
Primary investigator
- cjs
Secondary investigators
- ss03v
- yn05r
Partners
- Xerox Research Centre Europe
- Amebis
- Celer Soluciones
- Jozef Stefan Institute
- National Research Council Canada
- University of Bristol
- University of Helsinki
- Universita` degli Studi di Milano
- University College London
Associated research group
- Information: Signals, Images, Systems Research Group
- Date:
- 2006-
- Themes:
- Accessible Technologies, ELearning, Platforms and Tools, Learning Technologies
- Funding:
- IBM
RealTimeAnnotate will enable the highlighting and tagging of captions and the addition of synchronises notes.
Primary investigator
Partners
- IBM
- Liberated Learning Consortium
Associated research group
- Learning Societies Lab
- Date:
- 2006-
- Themes:
- Accessible Technologies, ELearning, Platforms and Tools, Learning Technologies
- Funding:
- IBM
RealTimeMerge facilitates the display of real time captions from multiple speakers with utterances identified by speakers' names
Primary investigator
Partners
- IBM
- Liberated Learning Consortium
Associated research group
- Learning Societies Lab
- Date:
- 2005-
- Themes:
- Accessible Technologies, Learning Technologies, ELearning, Platforms and Tools
- Funding:
- IBM
Recognition errors will inevitably sometimes occur with speech recognition and so an application is being developed to allow errors in synchronised captions to be corrected in real time.
Primary investigator
Partners
- IBM
- Liberated Learning Consortium
Associated research group
- Learning Societies Lab
- Date:
- 2005-
- Themes:
- Accessible Technologies, Learning Technologies, ELearning, Platforms and Tools
- Funding:
- IBM
Liberated Learningââ¬â¢s research has shown that although projecting captions onto a large screen in the classroom has been used successfully by students in many situations an individual personalised and customisable display may be preferable or essential if only a few short lines of text are shown in the window because of the pause and separator settings and the rest of the screen being taken up by other material (e.g. Power Point displays). An application is therefore being developed to provide users with their own personal display on their own wireless enabled systems customised to their preferences (e.g. font, size, colour, text formatting and scrolling).
Primary investigator
Partners
- IBM
- Liberated Learning Consortium
Associated research group
- Learning Societies Lab
- Date:
- 1999-
- Themes:
- Accessible Technologies, ELearning, Platforms and Tools, Learning Technologies
- Funding:
- IBM
The Liberated Learning Consortium is dedicated to advancing speech recognition technology and techniques to create and foster barrier-free learning environments to improve accessibility to information. Consortium partners work collaboratively to find solutions to challenges, generate new knowledge, test, research and develop speech recognition technologies and actively raise public awareness about the Liberated Learning concept.
Primary investigator
Partners
- Liberated Learning Consotium Members
- IBM
Associated research group
- Learning Societies Lab
- Date:
- 2007-
- Themes:
- Technology Enhanced Learning, ELearning, Platforms and Tools, Learning Technologies
- Funding:
- University of Southampton Student Centredness Fund
This project, with the help of students, aims to explore successful 'e-skills' and supporting strategies, including the use of assistive technologies, in a world of complex computer-based interactions introduced by e-learning and on-line social networks.
The project has now been extended with the support of the University of Southampton Student Centredness Fund.
Primary investigators
Secondary investigator
Associated research group
- Web and Internet Science
- Date:
- 2007-2012
- Themes:
- Human Computer Interaction, Semantic Web, Interaction, Web Science
- Funding:
- EPSRC, Royal Academy of Engineering
>Introducing doingPad
The doingPad project (doing is pronounced "doyng") is about looking at ways to better capture (and then find again) all the little bits (and bigger bits) of information that inform our lives - the phone number quickly scribbled down on a post-it, the time of a movie, the address of that web site and the big idea we had while talking with that person we only just met.
Getting these thoughts into a laptop is often a challenge: the percieved cost of entry - openning up a file, typing in text, naming the file, saving it, then remembering where it is - is usually too high to make the capture worth while. Then try to find the file later - perhaps months later. What was it called? what was it about?
DoingPad is reconsidering:
- the way we enter information into the computer to begin with
- what information can be associated with the info automatically (where was i? what was i doing? who was i with? what is this likely about?)
- how this new interaction and association can facilitate rediscovering that information
The Collaboration
This project is the first collaborative research work that is part of the Web Science Research Initiative between MIT and the University of Southampton.
Support
Initial support for the project has also been provided by the Royal Academy of Engineering (UK) and the EPSRC (UK).
Primary investigators
- monica schraefel
- Max Van Kleek, MIT
- Michael Bernstein, MIT
- David Karger, MIT
Partners
- Web Science Reseasrch Initiative
- MIT
Associated research group
- Intelligence, Agents, Multimedia Group
- Date:
- 2007-2008
- Themes:
- Assessment, ELearning, Platforms and Tools, Educational Enhancement
- Funding:
- JISC
This short assessment project is a collaboration between Cambridge, Southampton, the Scottish Qualifications Agency. It is taking an existing prototype, to develop the Minibix item banking system to meet JISC's requirements for a QTI v2-based item banking system capable of supporting both high-stakes private item banks and low-stakes item banks for sharing questions suitable for formative assessment.
The development work will be carried out by Cambridge in CARET during the first phase of the project (March 2007 to September 2007). The project will draw on additional use cases supplied by the SQA for private banks and Southampton University for public banks (drawing on experience with e3an). During the second phase (October 2007 to March 2008) the team at Southampton will carry out a case study of the use of Minibix for e3an and CARET will document their experiences in using the item bank to support TSA.
The resulting system will be available as an open source demonstrator with documentation enabling further development by the JISC community and beyond. The project will document and demonstrate a set of services suitable for the integration of item banking systems with authoring and delivery systems (in collaboration with the companion projects in the JISC programme).
Primary investigator
- Steve Lay, University of Cambridge
Secondary investigators
- saw
- hcd
- Mhairi McAlpine, Scottish Qualification Agency
Associated research group
- Learning Societies Lab
- Date:
- 2007-
This is a dummy project to enable topic specific lists of publications to be generated.
Primary investigator
Associated research group
- Science and Engineering of Natural Systems Group