Skip to content

A C++ implementation of the RIBES machine translation evaluation measure.

License

Notifications You must be signed in to change notification settings

rudhbaskaran/ribes-c

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

######################
# ribes-c            #
#   by Graham Neubig #
######################

This is a C++ port of the RIBES evalation metric (originally written in Python). You can find a description of the metric in the following paper:

Hideki Isozaki, Tsutomu Hirao, Kevin Duh, Katsuhito Sudoh, and Hajime Tsukada,
"Automatic Evaluation of Translation Quality for Distant Language Pairs",
Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing (EMNLP),
pp. 944--952 Cambridge MA, October, 2010
-- http://aclweb.org/anthology-new/D/D10/D10-1092.pdf

Please note that this program is not the official evaluation software, and any evaluations using RIBES should use the official Python package here:
http://www.kecl.ntt.co.jp/icl/lirg/ribes/

However, we have confirmed sentence-level accuracy between this package and RIBES 1.02.3 with **one exception**: In the original RIBES, full-width spaces are treated as delimiters, and in this package full-width spaces are treated as words. As a result, if your data contains full width spaces you might get slightly different results.

***********
Compliation

ribes-c requires that Boost be installed, with at least the 
boost_program_options library. Once the headers and library are installed,
just run "make," and the program will be compiled.

***********
License

The software is available under the GNU General Public License.
The original Python implementation is
    Copyright (C) 2011  Nippon Telegraph and Telephone Corporation
and this C++ port is
    Copyright (C) 2012  Graham Neubig

About

A C++ implementation of the RIBES machine translation evaluation measure.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published