Hot on HuffPost Tech:

See More Stories
AOL Tech

MIT Computer Deciphers Ugaritic, an Ancient Semitic Language

a tablet with ugaritic scriptThe big brains at MIT aren't just interested in solar cells, robot desk lamps and gesture interfaces; they're also applying their evolved understanding of tech to decipher ancient languages. The team of Regina Barzilay, an associate professor in MIT's Computer Science and Artificial Intelligence Lab, have created a computer program that successfully deciphered a chunk of Ugaritic, a dead Semitic script dating to the 14th and 12th centuries BCE in what is now Syria.

An important evolutionary transition from older cuneiform scripts, Ugaritic is an abjad -- an alphabet of consonants which requires the reader to supply the vowels. The first re-discovery of the language occurred in 1928, but took years to translate, whereas the software developed by Barzilay and company needed only a matter of hours to decipher the script.

But, in order to get the job done, the program makes a number of assumptions: that Ugaritic is closely related to the Hebrew script, and that it shares cognates with other ancient languages. And, while the researchers claim that the system was able to connect 60-percent of the known Hebrew-Ugaritic cognates and 29 of the 30 Ugaritic letters to the Hebrew script, they admit that their system will not replace human translators.

Still, the software opens the possibility of better translation programs; researchers note that Google Translate only works with 57 languages, while thousands of others have yet to be included. [From: MIT News]

Tags: computers, ComputerScienceAndArtificialIntelligenceLaboratory, DeadLanguages, language, mit, ReginaBarzilay, top, translation, ugaritic