UD tree manipulation and historical POS tagging

Christian Hardmeier

This repository contains a number of experiments that are grouped together because they depend on some common code to represent and manipulate Universal Dependency trees.

This should not be considered released software, and documentation is lacking. Please contact me if you're interested in using any of it.

The only component of general interest is the historical POS tagger in histpos/histpos.py. This is the code that was used for the experiments in the following paper:

Christian Hardmeier (2016). A Neural Model for Part-of-Speech Tagging in Historical Text. Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics, pp. 922-931. PDF