CNGL launch World Cup 'Twanslation' Service

John, Riona, Maria using CNGL's 'Twanslation' Service

CNGL researchers have developed a system to allow football fans follow World Cup tweets on Twitter in their own language. The CNGL Twanslator:World Cup 2010 project is being coordinated by Prof. Andy Way from the School of Computing in DCU (Track leader for Integrated Language Technologies / Machine Translation CNGL - DCU) and Dr Declan Dagger (CNGL TCD).

Natural boundaries exist within Twitter based on spoken languages - typically you are only connected to people who communicate in a language you understand. Twanslator WC 2010 is an attempt to filter the information streams on Twitter during the world cup into a number of different languages and create match summaries. CNGL is collaborating with the SFI CLARITY centre on aspects on this work.

The Twanslator:World Cup 2010 project forms part of the CNGL efforts into personalised multilingual social networking (codenamed 'myisle'). Headed up by Prof. Vincent Wade (Deputy Director & Track Leader for Digital Content Management - TCD) and Dr. Declan Dagger (Postdoctoral researcher in Digital Content Management - TCD) 'myisle' seeks to enable the "right time web" by applying CNGL research and technologies to create personalised and localised web experiences on the fly.

Localisation and personalisation are seen as key enablers to break down the natural language barriers that exist in Social Media and filter the real-time information streams so that they are more relevant to each individual, i.e. towards "the right time web"

CNGL are running the DCU MT system OpenMaTrEx in real time in an online, on-demand application and provide a service (multilingual Tweet translation) where there is currently no provision. CNGL will engage the community in rating the translation output. Incremental improvement will be measures as WC 2010 progresses.

The languages covered are English to French, German, Spanish, Portuguese, Dutch, Italian - and FR, DE, ES, PT, NL, IT to EN.

A dedicated webpage is running at http://www.myisle.org/worldcup2010. This allows the user to select their language preference for the information stream and displays a rolling digest of the tweets from the world cup (translated into their selected language). The page also illustrates, using sentiment analysis techniques, a summary of events during the match.

The translated streams are pushed back onto the twitter network so that people can access the translated streams through their favourite twitter client. To achieve this CNGL have set up specific Twitter accounts with language identifiers, {e.g. twc2010_en, twc2010_fr, twc2010_es, twc2010_it, twc2010_de, twc2010_it, twc2010_pt, twc2010_nl}, which people can follow.

You can view the front page Sunday Time article here: http://www.timesonline.co.uk/tol/news/world/ireland/article7144941.ece

CNGL is happy to announce we have two new contract positions available to work in the area of personalised multilingual social networking at Trinity College. Please see http://www.cngl.ie/vacancies.html for descriptions of the positions