{"id":95187,"date":"2020-10-22T17:12:50","date_gmt":"2020-10-22T14:12:50","guid":{"rendered":"https:\/\/en.buradabiliyorum.com\/new-mit-algorithm-automatically-deciphers-lost-languages\/"},"modified":"2020-10-22T17:12:50","modified_gmt":"2020-10-22T14:12:50","slug":"new-mit-algorithm-automatically-deciphers-lost-languages","status":"publish","type":"post","link":"https:\/\/buradabiliyorum.com\/en\/new-mit-algorithm-automatically-deciphers-lost-languages\/","title":{"rendered":"#New MIT algorithm automatically deciphers lost languages"},"content":{"rendered":"

“#New MIT algorithm automatically deciphers lost languages<\/strong>”
\n<\/p>\n

\n A new AI system can automatically decipher a lost language that\u2019s no longer understood \u2014 without knowing its relationship to other languages.<\/p>\n

Researchers at MIT CSAIL developed the algorithm in response to the rapid disapp<\/a>earance of human languages. Most of the\u00a0languages that have existed are no longer spoken, and at least half of those remaining are predicted to vanish in the next 100 years<\/a>.<\/p>\n

The new system could help recover them. More importantly, it could preserve our understanding\u00a0of the cultures and wisdom of their speakers.<\/p>\n

[Read:\u00a0What audience intelligence data tells us about the 2020 US presidential election]<\/span><\/em><\/p>\n

The algorithm works by harnessing key principles from historical linguistics, such as the predictable ways in which languages use sound substitutions. The researchers give the example of a word with a \u201cp\u201d in a parent language possibly changing to a \u201cb\u201d in its descendent, but probably not to a \u201ck\u201d due to the difference in pronunciation.<\/p>\n

These kinds of patterns are then turned into computational constraints. This allows the model to segment words from an ancient language and map them to a related language.<\/p>\n

The algorithm can also identify different language families. For example, their method suggested that Iberian is not related to Basque, supporting recent scholarship.<\/span><\/p>\n

The project was led\u00a0by MIT Professor Regina Barzilay, who last month won a $1 million award<\/a> from the\u00a0world\u2019s largest AI<\/span>\u00a0society\u00a0for her pioneering work on drug development and breast cancer detection.<\/span><\/p>\n

She now wants to expand the work to identifying the semantic meaning of words \u2014 even if we don\u2019t know how to read them.<\/p>\n

\u201cFor instance, we may identify all the references to people or locations in the document which can then be further investigated in light of the known historical evidence,\u201d Barzilay said in a statement.<\/p>\n

\u201cThese methods of \u2018entity recognition\u2019 are commonly used in various text processing applications today and are highly accurate, but the key research question is whether the task is feasible without any training data in the ancient language.\u201d<\/p>\n

You can read the full research study here<\/a>.<\/p>\n

\n Published October 22, 2020 \u2014 14:12 UTC\n <\/p>\n<\/p><\/div>\n