Paper: Simpler unsupervised POS tagging with bilingual projections

ACL ID P13-2112
Title Simpler unsupervised POS tagging with bilingual projections
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2013

We present an unsupervised approach to part-of-speech tagging based on projec- tions of tags in a word-aligned bilingual parallel corpus. In contrast to the exist- ing state-of-the-art approach of Das and Petrov, we have developed a substantially simpler method by automatically identi- fying ?good? training sentences from the parallel corpus and applying self-training. In experimental results on eight languages, our method achieves state-of-the-art re- sults. 1 Unsupervised part-of-speech tagging Currently, part-of-speech (POS) taggers are avail- able for many highly spoken and well-resourced languages such as English, French, German, Ital- ian, and Arabic. For example, Petrov et al. (2012) build supervised POS taggers for 22 languages us- ing the TNT tagger (Brants, 2000), with an aver- a...