NLP for Endangered Languages: Morphology Analysis, Translation Support and Shallow Parsing of Ainu Language
Michal E. Ptaszynski, Mukaichi Kazuki, Yoshio Momouchi · 2013
This paper describes our research on computer processing of Ainu language with the use of various NLP techniques. Ainu is an endangered language close to extinction. At present linguists and anthropologists make a great effort to preserve the language by analyzing and understanding it. However, most of the work in this matter is done manually, which makes it an uphill task. Previously we have presented POST-AL, a part-of-speech tagger for Ainu language. This paper describes recent improvements to the system as well as other enhancements made with an aim to help Ainu language researchers. In particular, we have enhanced the POS tagger with analysis of morphological information. We have also added a translation support tool for Ainu language translators and made a first step toward deeper syntactical analysis of Ainu language by creating a simple shallow parser.