Comments on: Language Weaver CTO Says Improvements to Statistical Machine Translation Expand Opportunities for Government Customers http://blog.languageweaver.com/index.php/2008/07/09/language-weaver-cto-says-improvements-to-statistical-machine-translation-expand-opportunities-for-government-customers/ Language Weaver Blog Tue, 07 Sep 2010 00:23:37 +0000 http://wordpress.org/?v=2.7.1 hourly 1 By: Anton http://blog.languageweaver.com/index.php/2008/07/09/language-weaver-cto-says-improvements-to-statistical-machine-translation-expand-opportunities-for-government-customers/comment-page-1/#comment-132 Anton Fri, 11 Jul 2008 14:44:34 +0000 http://blog.languageweaver.com/2008/07/09/language-weaver-cto-says-improvements-to-statistical-machine-translation-expand-opportunities-for-government-customers/#comment-132 Hello! Please forgive me if this post is inappropriate, but I couldn't find a direct email address on your blog. I'm Anton and I'm launching my new blog dealing with language translation issues and would appreciate the opportunity to discuss mutual collaboration. Thanks! Hello! Please forgive me if this post is inappropriate, but I couldn’t find a direct email address on your blog. I’m Anton and I’m launching my new blog dealing with language translation issues and would appreciate the opportunity to discuss mutual collaboration. Thanks!

]]> By: Adam Zand http://blog.languageweaver.com/index.php/2008/07/09/language-weaver-cto-says-improvements-to-statistical-machine-translation-expand-opportunities-for-government-customers/comment-page-1/#comment-129 Adam Zand Wed, 09 Jul 2008 18:24:21 +0000 http://blog.languageweaver.com/2008/07/09/language-weaver-cto-says-improvements-to-statistical-machine-translation-expand-opportunities-for-government-customers/#comment-129 Nice job Daniel and Tim. It would be interesting to hear a bit more about how governments use machine translation. I wonder if it is enough to "pipe in" a feed from a monitoring station in a hot spot or do you have to "tune" the engine to teach it code word and references that might seem harmless. For example, in the Jose Padilla case the prosecution explained the absence of any terrorism discussion in recorded conversations by claiming that the defendants were speaking in code. Thus "soccer equipment"supposedly meant guns, "eating cheese" meant violent jihad, etc. How does machine translation keep up with the changing code? Nice job Daniel and Tim. It would be interesting to hear a bit more about how governments use machine translation.

I wonder if it is enough to “pipe in” a feed from a monitoring station in a hot spot or do you have to “tune” the engine to teach it code word and references that might seem harmless.

For example, in the Jose Padilla case the prosecution explained the absence of any terrorism discussion in recorded conversations by claiming that the defendants were speaking in code. Thus “soccer equipment”supposedly meant guns, “eating cheese” meant violent jihad, etc.

How does machine translation keep up with the changing code?

]]>