Abstract
We report on the evaluation of the Norwegian–English MT prototype system LOGON. The system is rule-based and makes use of well-established frameworks for analysis and generation (LFG and HPSG). Minimal Recursion Semantics is the "glue" which performs transfer from source to target language and serves as the information vehicle between LFG and HPSG. The project-internal testing uses material from the training data sources. We report on two test methods in addition: 1) on test data from the same (narrow) domain as the project was designed for, and 2) on a syntactic test suite. The former turns out to give much worse results than the project-internal tests, while the latter obtains somewhat better results. These results are important for future evaluations of similar systems.
Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC'08), pages 396-402. Published with the permission of ELRA. This paper was published within the proceedings of the LREC 2008 and LREC 2012 Conferences. © 2008-2012 ELRA - European Language Resources Association. All rights reserved.