Create POS Tagged and Parsed Text in Senseval-2 format
These tools are intended for use with the
Brill Part of Speech Tagger
and the
Collins Parser. Their objective is to part of speech tag and parse
text that is stored in the Senseval-2 format.
parseSenseval
This package allows a user to parse any text in the Senseval-2 format
using the Collins Parser. The text should have already been POS tagged
(using posSenseval).
posSenseval
This package allows a user to part of speech tag any text in the
Senseval-2 format with the Brill POS Tagger.
Guaranteed Pre-Tagging
The user may optionally apply the following patch to the Brill tagger
in order to allow the user to specify a pre-tag for a word and guarantee
that it be used by the tagger. This patch is incorporated into the
posSenseval package if the user has it installed.
Pre-Processing for Tagging
This package will format the Senseval-2 data appropriately for the Brill
Tagger, and allow for the easy insertion of pretags.
By:
Ted Pedersen
- tpederse AT d umn edu