From Taggers to Parsers
Pam Peters
School of English, Linguistics and Media, Macquarie University
When: Tuesday, 6th December 1995
Time: 4:00pm
Where: Room E6A357, Macquarie University
Abstract:
This session will take a long look at the output of Eric Brill`s tagger, which we have been applying to various types of natural speech collected for the ICE database (= International Corpus of English). The tagger recommends itself as one which is robust and trainable, and the session is intended to examine its accuracy and reliability on speech material. We shall also consider ways of developing it to overcome certain problems (some of these relate to spoken language, but others show up in written language as well). Self-contained parsers such as ALVEY Natural Language Tool and Susanne will also be reviewed, as well as Tosca, intended for the ICE material.
Enquiries: Maria Milosavljevic 9850 6345 mariam@mpce.mq.edu.au
| Last modified: July 1997 |