BLEU: a method for automatic evaluation of machine translation
ACL '02 Proceedings of the 40th Annual Meeting on Association for Computational Linguistics
| Hi-index | 0.00 |
This paper presents an idea in Example-Based Machine Translation - computing the transfer score for each produced translation. When an EBMT system finds an example in the translation memory, it tries to modify the sentence in order to produce the best possible translation of the input sentence. The user of the system, however, is unable to judge the quality of the translation. This problem can be solved by providing the user with a percentage score for each translated sentence. The idea to base transfer score computation on the similarity between the input sentence and the example is not sufficient. Real-life examples show that the transfer process is as likely to go well with a bad translation memory example as to fail with a good example. This paper describes a method of computing transfer score strictly associated with the transfer process. The transfer score is inversely proportional to the number of linguistic operations executed on the example target sentence. The paper ends with an evaluation of the suggested method.