|
| Speaker: | |
| Norwegian Computing Centre for the Humanities, University of Bergen | |
| Date: | 14th April 1998 |
| Time: | 11:30am |
| Place: | Seminar room 357, Building E6A, Department of Computing, Macquarie University |
Abstract:
Trained on annotated corpora, parsers can produce outputs of syntactic analysis on unseen sentences based on the annotations of the training corpus. In this talk, I will introduce T-tag Oriented Parsing, a corpus-based approach that 1) sees the annotations in corpora as structural boundary labels (T-tags), 2) retrieves T-tags while the parser is trained, and 3) parses new text with the T-tags. I will also discuss heuristic techniques applied in the parsing.
Enquiries: sals@mri.mq.edu.au
| Last modified: March, 1998 |