Paper: Minimized Models for Unsupervised Part-of-Speech Tagging

ACL ID P09-1057
Title Minimized Models for Unsupervised Part-of-Speech Tagging
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2009
Authors

We describe a novel method for the task of unsupervised POS tagging with a dic- tionary, one that uses integer programming to explicitly search for the smallest model that explains the data, and then uses EM to set parameter values. We evaluate our method on a standard test corpus using different standard tagsets (a 45-tagset as well as a smaller 17-tagset), and show that our approach performs better than existing state-of-the-art systems in both settings.