Seminarier och gästföreläsningar
Stockholm TreeAligner - A tool for aligning and searching parallel treebanks
The Stockholm TreeAligner allows you to create alignment links between corresponding nodes (or words) in two treebanks in different languages. This is useful for creating parallel treebanks, which have a wide array of linguistic applications, most notably machine translation.
The Stockholm TreeAligner displays trees from input files in TigerXML format with node labels, edge labels, and even crossing branches, making it useful for browsing TigerXML files as well.
As of version 0.8 (13. Dec. 2007) the Stockholm TreeAligner allows querying parallel treebanks (inspired by the TIGERSearch query language). Search results are highlighted in a graphical display.
This program is free software.
You may redistribute it and/or modify it under the terms of the GNU General Public License as published by the Free Software Foundation; either version 2 of the License, or (at your option) any later version.
The following screen shot shows a German tree and an English tree of the corresponding sentence. The nodes are aligned across the trees. Green lines stand for 'exact' translations while red lines stand for 'approximate' translations. The example contains node alignments and word alignments. We allow 1:n alignments (as e.g. in the alignment of German 'Heimweg' to English 'way home').
... is described on the TreeAligner Wiki page
... is described in the TreeAligner User Manual which is part of the delivery.
Stockholm TreeAligner is a project of the Computational Linguistics Groups at Stockholm University and the University of Zurich
Project head: Martin Volk, Email: email@example.com
Contributors: Joakim Lundborg, Torsten Marek, Maël Mettler, Sandra Roth
We gratefully acknowledge support by Granholmsstiftelsen 2005 and by the IIL program of the University of Zurich in 2009.
This software is a tribute to Maël Mettler who died unexpectedly in October 2007. His contributions to the software (the query module and indexing schemes) but even more his enthusiasm for our project will always be remembered. We have lost a friend.
27 april 2011
Webbredaktör: Ulla Bjursäter
Sidansvarig: Department of Linguistics/Section for Computational Linguistics