TrebleCLEF Logo
Evaluation, Best Practices and Collaboration for Multilingual Information Access
Latest News
CLEF 2010: Padua, Italy September 2010
Read more - Link

TrebleCLEF workshop at eChallenges : Best Practices for Multilingual Information Access Istanbul,...
Read more - Link

Read all news

CMU makes available OpenEphyra open source QA system.

The question answering research group at CMU has decided to place the Ephyra QA system into open source. The open source version - OpenEphyra - is almost identical to the system that has been evaluated in the QA track at TREC, except that  some 3rd party tools and code with specific hardware requirements have been excluded. The result is a system that is platform-independent, easy to use, and that can be run on a standard desktop computer.

You are invited to download OpenEphyra and try it out for your own research.
By making the code open, CMY hopes to facilitate evaluations and comparisons of different approaches by providing a common platform for experiments. In addition, they would like to give researchers the opportunity to develop new algorithms for QA processing steps without having to build an entire end-to-end system first. It is their hope that OpenEphyra will also be used for educational purposes; it could easily be adapted for computer science course projects on NLP applications.

If you are interested in trying it out, you can download OpenEphyra from
Just unzip it and run one of the scripts.

Additional information about the project, the documentation and a collection of tutorials can be found here:

Also, see the mailing list on which  future releases and news about OpenEphyra will be announced:
This is a low-traffic list.

Finally, we would like to invite you to get involved in this open QA project. We highly encourage you to extend the system with your own work, and to share your improvements with us and the QA community. 
Please get in touch with us if you are interested in joining this project, or if you have any comments or questions!

Information supplied by:

Nico Schlaefer
PhD candidate
School of Computer Science
Carnegie Mellon University