I have a prototype of an interactive sentence aligner and another one
for interactive word alignment using web-based interfaces. They are
part of Uplug (http://sourceforge.net/projects/uplug/) and you may try
them with small sample texts at
http://www.let.rug.nl/~tiedeman/__align_test/isa.php
(interactive sentence aligner)
http://www.let.rug.nl/~tiedeman/__align_test/ica.php
(interactive word aligner)
(Be nice to our server! - I might take the scripts away if it
starts to get too busy! - But I thought I might risk it to put them
on-line for a short while. I have a ulimit of 5 seconds for running
alignments. Don't get confused if it times out ...)
Current features of the sentence aligner:
- automatic alignment with Gale&Church's approach
(click on 'align')
- a simple cognate filter to add hard-boundaries
(press the 'cognates' button)
- adding hard-boundaries using structural (XML) tags
(there are no useful ones in the test corpus)
- manually adding/removing hard-boundaries
(click on corresponding sentences)
- saving/sending results in 3 different formats
(XCES with external pointers, simple TMX, plain text)
Current features of the word aligner:
- uses the Uplug clue aligner (with existing clue collections)
- alignment visualization (matrix style)
- select clue types and weights
- add/remove links (click on corresponding cells in the matrix)
- open/search the clue collections
- save alignment results (disabled in the demo)
More information can be found at
http://www.let.rug.nl/~tiedeman/__align_test/doc/isa.html
http://www.let.rug.nl/~tiedeman/__align_test/doc/ica.html
I'm looking forward to get your feedback!
best,
Jörg Tiedemann
***********/\/\/\/\/\/\/\/\/\/\/\************************************
** Jörg Tiedemann tiedeman@let.rug.nl **
** Alfa-Informatica http://www.let.rug.nl/~tiedeman **
** Rijksuniversiteit Groningen Harmoniegebouw, room 1311-429 **
** Oude Kijk in 't Jatstraat 26 phone: +31 (0)50-363 5935 **
** 9712 EK Groningen fax: +31 (0)50-363 6855 **
*************************************/\/\/\/\/\/\/\/\/\/\/\**********
This archive was generated by hypermail 2b29 : Thu Jul 28 2005 - 18:00:03 MET DST