• Overview
• DependencyVectors
• sigf
• German NER

DependencyVectors

DependencyVectors is a Semantic Space model construction program that works on the basis of Dependency Grammar relations.
Currently, it only works on corpora parsed with the MINIPAR dependency parser, but it should be easily adaptable to the output of other parsers. It is written in Java and requires a Java 1.5 runtime environment. (If you need a 1.4 version, write me an email.)

Download

DependencyVectors is available under the terms of the GNU General Public License.

• Download DependencyVectors 2.4a (sources and documentation) as a tgz archive (about 100K)
• Download the DependencyVectors 2.4a Manual as pdf file

History

• V2.4a (November 2009): Bugfixes, refactoring, preparations
• V2.4 (July 2008): Bugfixes, Java 1.5, modular Basis Mapping
• V2.3 (January 2008): Bugfixes
• V2.2 (August 2007): More predictable string handling
• V2.1a (July 2007): Minor bugfix; added sample
• V2.1 (May 2007): Class interfaces now more consistent and easier to extend
• V2.0a (August 2005): Minor bugfix -- javadoc failed with Java 1.5
• V2.0 (December 2004): Reimplementation

Usage

Usage is explained in the manual you can download above.

Publications

• S. Pado and M. Pennacchiotti and C. Sporleder: Towards semantic role labelling for event nominalisations through bootstrapping from verbal data. Proceedings of COLING 2008. Manchester, UK.
• S. Pado and U. Pado and K. Erk: Flexible, Corpus-Based Modelling of Human Plausibility Judgements . Proceedings of EMNLP-CoNLL 2007, Prague.
• S. Pado and M. Lapata: Dependency-based Construction of Semantic Space Models. Computational Linguistics 33(2), 2007.
• S. Pado and M. Lapata: Constructing Semantic Space Models from Parsed Corpora. Proceedings of ACL-03, Sapporo.

Contact

Feedback is always welcome at pado%40coli.uni-sb.de.