Using Machine Learning to Translate Long-Lost Languages

Though languages evolve over time, they don't necessarily undergo dramatic changes when it comes to the symbols used or distribution of words and characters. And it would be possible to decode an unknown language if it's related to a known progenitor language.

Based on the premise, a team from MIT have used machine learning to crack certain languages which could be used as a technique that would allow us to decode even long-lost languages, by sheer brute force. The difference being, machines don't get fatigued and have copious databases with which to compare texts and symbols.

Enter Jiaming Luo and Regina Barzilay from MIT and Yuan Cao from Google’s AI lab in Mountain View, California. This team has developed a machine-learning system capable of deciphering lost languages, and they’ve demonstrated it by having it decipher Linear B—the first time this has been done automatically.
The idea is that any language can change in only certain ways—for example, the symbols in related languages appear with similar distributions, related words have the same order of characters, and so on. With these rules constraining the machine, it becomes much easier to decipher a language, provided the progenitor language is known.

(Image credit: Tilemahos Efthimiadis/Flickr; Wikimedia Commons)


Login to comment.




Email This Post to a Friend
"Using Machine Learning to Translate Long-Lost Languages"

Separate multiple emails with a comma. Limit 5.

 

Success! Your email has been sent!

close window
X

This website uses cookies.

This website uses cookies to improve user experience. By using this website you consent to all cookies in accordance with our Privacy Policy.

I agree
 
Learn More