네이처의 실린 기사, “AI가 고대 문자를 해독하고 역사를 다시 쓰는 방법”. 아래는 한국 사료 사례.
“South Korean researchers, meanwhile, are facing very different challenges as they tackle one of the world’s largest historical archives: detailed daily records with hundreds of thousands of articles covering the reigns of 27 Korean kings, dating from the fourteenth to the early twentieth centuries. “The amount of data is vast,” says Kyunghyun Cho, a leading machine-translation researcher at New York University in New York City. Cho usually works with modern languages, but became interested in the archives after discussing them with his father, a retired professor of Korean literature. These records are complete and their origins are known, but hardly anyone can read them. They are written in Hanja, an ancient writing system based on Chinese characters that is different from modern Chinese or Korean.”
https://www.nature.com/articles/d41586-024-04161-z
