[Corpora-List] Shallow semantic parser - New release

Sebastian Pado pado at CoLi.Uni-SB.DE
Tue Mar 13 16:48:01 CET 2007


Dear all,

This is to announce the release 1.1 of Shalmaneser (a SHALlow seMANtic
parSER), a system for automatic sense assignment and semantic role
labeling. Shalmaneser comes with pre-trained FrameNet classifiers for
English and German. (Further details below.)

Improvements over the previous release (1.0) include:

- Pre-trained English classifiers for FrameNet release 1.3
(i.e., improved coverage and accuracy)
- Support for TreeTagger as POS Tagger for both English and German
- Removal of numerous bugs

You can download the software from this URL:

http://www.coli.uni-saarland.de/projects/salsa/shal/

Again, please let us know if you have any questions, comments, or
suggestions!

Best,

Sebastian Pado and Katrin Erk

---
Purpose of Shalmaneser

Shalmaneser has been developed with two uses in mind: research in
applications that use shallow semantic parses, and research on better
shallow semantic parsing. So Shalmaneser can be used either as a
'black box' to obtain semantic parses for free text, or as a research
platform that can be extended to new parsers, languages, or
classification paradigms.


Features of Shalmaneser

- Shallow semantic parser: word sense disambiguation for predicates,
plus semantic role labeling
- Input: plain text. Syntactic processing integrated.
- Classifiers available: trained on FrameNet data for English and
German (System also applicable to other frameworks)
- System output can be viewed graphically in the SALTO viewer:
http://www.coli.uni-saarland.de/projects/salsa/page.php?id=software
- System realized as a toolchain of independent modules communicating
through a common XML format, hence extensible by further modules
- Interfaces for addition/exchange of parsers, learners, features


More information

K. Erk and S. Pado: Shalmaneser - a flexible toolbox for semantic role
assignment. Proceedings of LREC-06, Genoa.
http://www.coli.uni-saarland.de/~pado/pub/papers/lrec06_erk.pdf





More information about the Corpora-archive mailing list