Asiya - Online
An Online Toolkit for Automatic Machine Translation Evaluation
- TOOL: ASIYA Online
- AUTHOR(s): Jesús Giménez, Meritxell Gonzàlez
- DATE: March 2013
- VERSION: 1.0
- DESCRIPTION: An Online Toolkit for Automatic MT Evaluation
Asiya is a software for Automatic Machine Translation (meta-)evaluation.
Asiya-Online allows accessing the Asiya Framework for testing purposes, with no need to install packages. - Asiya Forum and News
Discussion on features and bugs of this software as well as information about updates takes place on the Asiya News Feed to which you can subscribe at:
* http://nlp.cs.upc.edu/redmine/projects/asiya/news.atom
and post messages at the Forum:
* http://nlp.cs.upc.edu/redmine/projects/asiya/boards
Any feedback (comments, suggestions, bug reporting) from the nlp community members will be highly appreciated. - CONTRIBUTING
The Asiya package is licensed under LGPL, which means that it may be linked to and used by commercial software packages. But the license also enforces that any changes or improvements made to the library must be redistributed under LGPL terms.
Thus, if you improve the software or data, either adding new functionalities, fixing bugs, or adding new metrics, you can not distribute them under different conditions than those stated in the license (i.e. freely and with no usage restrictions).
Instructions:
- Select the input type of your data files (xml|text)
- Select the source language and case
- Select the target language and case
- Upload the files (max. 200K each, UTF-8 Unicode):
- one sentence per line.
- all files MUST have the same number of lines.
- one source text file.
- at least one, and up to five, candidate translations.
- at least one, and up to five, reference translations.
- Select the subset of candidate and reference translations to evaluate (default: all).
- Select the metrics (default selection is provided).
- Click on the "Run Asiya!" button.
- Execution should take from one to five minutes for English, and a just a few seconds for other languages (no linguistic processors are used).