Greek Translation and the Web

Tagged: Machine Translation RSS Toggle Comment Threads | Keyboard Shortcuts

• Dimitris 12:41 on 27/01/2011 Permalink
  Tags: Machine Translation   

  Intensive School in Machine Translation 

  

  Joint EAMT – SUSU Intensive School in Machine Translation

  European Association for Machine Translation and South Ural State University are pleased to announce an exciting co-sponsored event “Intensive School in Machine Translation (ISMT)” which will be held in Chelyabinsk, Russia from 16 to 20 May 2011. The venue of the school is South Ural State University.

  The overall goal of ISMT is to provide participants with a reference frame for major areas of study within the field of MT technology while defining critical problems of MT and the most useful methods for their solutions.

  The ISMT courses on the state-of-the-art MT aspects will be delivered by internationally renowned researchers from different countries. Click here to see the lecturers.

  ISMT will also feature a discussion panel which will enable participants to discuss specific issues in MT with experts in the field. The working languages are English and Russian. A course certificate will be issued upon completion of the course, based on 90% participation.

  The primary ISMT target audience is university students, PhD students and young researchers for whom attendance is free of charge. The school is also open for everyone else professionally interested in machine translation. Participants from industry/commerce will be charged a registration fee.

  Since there are a limited number of participant slots we do recommend interested parties to register as early as possible. Assistance with accommodation is offered once you have registered.

  Important Dates
  15 January 2011		Registration opens    Click here to register
  15 April 2011		Registration deadline
  16-20 May 2011		Intensive School in Machine Translation

  Contact: org@mtschool-susu.info

  The ISMT courses on the state-of-the-art MT aspects will be delivered by internationally renowned researchers from different countries. Continue reading “Intensive School in Machine Translation” »
   

• Dimitris 21:46 on 22/11/2010 Permalink
  Tags: Machine Translation, medical translation   

  Article: Machine Translation-Supported Cross-Language Information Retrieval for a Consumer Health Resource 

  Article Title: Machine Translation-Supported Cross-Language Information Retrieval for a Consumer Health

  Authors: Graciela Rosemblat, Darren Gemoets, Allen C. Browne, and Tony Tse | National Library of Medicine, Maryland

  Abstract:

  The U.S. National Institutes of Health, through its National Library of Medicine, developed ClinicalTrials.gov to provide the public with easy access to information on clinical trials on a wide range of conditions or diseases. Only English language information retrieval is currently supported. Given the growing number of Spanish speakers in the U.S. and their increasing use of the Web, we anticipate a significant increase in Spanish-speaking users. This study compares the effectiveness of two common cross-language information retrieval methods using machine translation, query translation versus document translation, using a subset of genuine user queries from ClinicalTrials.gov. Preliminary results conducted with the ClinicalTrials .gov search engine show that in our environment, query translation is statistically significantly better than document translation. We discuss possible reasons for this result and we conclude with suggestions for future work.

  Source: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1480254/

  Download full article on pdf: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1480254/pdf/amia20030564.pdf

  Continue reading “Article: Machine Translation-Supported Cross-Language Information Retrieval for a Consumer Health Resource” »
   

• Dimitris 10:17 on 19/11/2010 Permalink
  Tags: Machine Translation, post editing   

  Machine Translation Post-Editing: Bring it on! 

  Following SDL’s announcement of its new Cloud platform for automated translation (BeGlobal), I couldn’t help but think about the issue of “Quality”. Where does Quality fit in to this new era in our profession? Up until now, it was one of the three main pillars of the translation industry (the others being cost and time). According to SDL, users of this new cloud system will select their preffered quality level and If the result does not reach the required level, the relevant content will be sent to post-editors. Fair enough…

  The professional translator/post-editor (by the way, we should add this skill to our business cards, it sounds fancy! ) who is appointed for such a task, will have to determine the amount of time and effort to put on this project. It’s fair to assume that the rate will be 1/2 or 1/3 of the usual, so the translator will have to allocate a carefully calculated amount of time and effort if he/she wants to make a reasonable wage. But how is he/she supposed to do that?

  We all know (and been there) that some translators, especially beginners, tend to treat every text as their “child” and spend a lot of time and effort on it. Of course, after a year or so they end up having more children than Angelina Jolie (but that’s another issue). On the other hand, there is another school of thought that says a translator should work on a text according to the amount he is getting paid for it. This amount in the case of post-editing will not be significant, but the task itself is not the easiest one. Hence, either the translator will spend endless hours trying to form a good quality text just because his business ethics do not allow him/her to deliver a half-good text, or the translator will just move a few words here and there to make the text a bit more readable (which is what post-editing is all about in the first place – but I am not sure If all parties are aware of that).

  Each one of these scenarios has its consequences: The “diligent” translator will probably have to look for a night job and the “professional” translator will have problems getting as many post-editing jobs by those agencies or clients that think post-editing equals re-writing. Because, let’s face it, there are going to be agencies and companies that will try to fool (mostly) young translators into believing that they have to achieve the same quality as before but with half the rates. This clearly shows that during this transition period, clients and agencies need to become familiar with this new platform, its nature and its capabilities, and translators need to have their eyes open for scammers.

  Continue reading “Machine Translation Post-Editing: Bring it on!” »
   

• Dimitris 13:17 on 15/11/2010 Permalink
  Tags: Content creation, Machine Translation   

  WikiBhasha: A new multi-lingual content creation tool for Wikipedia 

  WikiBhasha beta is a browser-based application that helps a community of users to create multilingual content in their Wikipedia, leveraging existing content from other Wikipedias. This application manifests as a simple and intuitive user interface layer that stays on the target language Wikipedia article that is being created or enhanced through this application, for the duration of the user session. At the end of the session, all content created or modified are submitted to the target language Wikipedia.

  The UI layer implements a simple 3-step process helping the user in (i) gathering content from source language Wikipedia, (ii) composing the target language Wikipedia article, and finally, (iii) submitting new content to the target Wikipedia. The UI layer integrates content discovery, linguistic and collaborative services, transparently. WikiBhasha beta’s workflow may be customized as per the requirements of specific user communities. Also, customization is possible at many levels, including adding support for many 3rd party services or for handling Wikipedia specific issues.

  Though WikiBhasha application is architected to be transparent with respect to languages of the source and target Wikipedias, the beta version focuses on leveraging the large English Wikipedia content primarily, for enhancing the content in non-English Wikipedias. Hence, WikiBhasha beta currently supports all translation pairs in Bing Translator, in which the source language is English. WikiBhasha beta currently tested on Internet Explorer (7 or above) on Windows XP, Windows Vista & Windows 7 and on Firefox (3.5 or above) on Windows and Linux (Fedora 11 or above).

  Source: http://en.wikipedia.org/wiki/User:Kit_Lywait/WikiBasha
  More info: Link
   

• Dimitris 12:22 on 14/11/2010 Permalink
  Tags: Machine Translation, Social media   

  Crowdsourcing Translation 

  

  “Crowdsourcing is the act of outsourcing tasks, traditionally performed by an employee or contractor, to a large group of people or community (a crowd), through an open call.” Wikipedia

  “Translation Crowdsourcing” stepped into the spotlight back in 2007-2008 when Facebook decided not to engage professional translators in the localization of its content into different languages, but instead involve its immense user community. This “open call” received many enthusiastic responses and the translation result, although dodgy in some cases, was really not that bad. Of course one can easily spot the occasional inconsistent capitalization, non-localized punctuation and untranslated legal terms…

  Facebook created a Translation app which, in their own words, worked as follows: “Facebook’s Translation app presents users with words that need to be translated, and they submit their entries. The system then invites other users to vote using Reddit-style up/down arrows to vote on which translations are best.” Next thing you know, FB had a localized version in more than 60 languages, for free!

  As for the ethics of this process, it has been argued that it stands on the borders of legitimacy. I can’t help but thinking that a 15 year old boy from Philippines is localizing content (for free) for a multi-billion dollar business. Isn’t the result of his labour bringing “more” clients to this business? What’s in it for him or his country anyway? These questions are reasonable, although a bit exaggerated (including my comment). Crowdsourcing is here to stay and If the very fact that you enjoy a web platform makes you want to help localize it, then so be it. Why not? Of course this huge corporation could reimburse you in some way for your labour, but online communities do not really work like this, do they?

  Crowdsourcing is turning into a form of Machine Translation by humans (peculiar as it may sound). I am not sure which one is better in terms of quality, as both non-professional human translators and machine translation systems tend to translate word for word. Hence I believe it’s safe to assume that it’s not likely for crowdsourcing to affect Medical, Legal, Literary or other more demanding forms of translation work, since the added quality of professional and specialized translators will always be necessary. But it will play a role (hand in hand with machine translation) in the localization of web communities, open-source software, web applications, charity and activism ….everything that gains public interest or sympathy or is just …”hip and happening”!

  French Translation Edinburgh is discussing. Toggle Comments

   

  • French Translation Edinburgh 23:16 on 28/01/2011 Permalink | Log in to Reply

    I think they are just looking for free labour. I would advise people to be very careful on who they offer their time and effort for.