Computational strategies for reducing annotation effort in language documentation

With the urgent need to document the world’s dying languages , it is important to explore ways to speed up language documentation efforts. One promising avenue is to use techniques from computational linguistics to automate some of the process. Here we consider unsupervised morphologica l segmentation and active learning for creating interlinear glossed text ( IGT) for the Mayan language Uspanteko. The practical goal is to produce a total ly annotated corpus that is as accurate as possible given limited time for man ual annotation. We discuss results from several experiments that suggest th ere is indeed much promise in these methods but also show that further developm ent is necessary to make them robustly useful for a wide range of conditions an d t sks. We also provide a detailed discussion of how two documentary lingui sts perceived machine support in IGT production and how their annotation p erformance varied with different levels of machine support. 1 LiLT Volume X, Issue Y, November 2009. Computational strategies for reducing annotation effort i n language documentation . Copyright c © 2009, CSLI Publications. 2 / L I LT VOLUME X, ISSUEY NOVEMBER 2009