Sensitivity of Automated MT Evaluation Metrics on Higher Quality MT Output: BLEU vs Task-Based Evaluation Methods.
Bogdan BabychAnthony HartleyPublished in: LREC (2008)
Keyphrases
- evaluation metrics
- evaluation methods
- higher quality
- machine translation
- precision and recall
- high quality
- evaluation methodology
- information extraction
- cross lingual
- natural language processing
- evaluation measures
- reconstructed image
- statistical machine translation
- cross language information retrieval
- evaluation method
- learning to rank
- natural language
- machine learning
- machine translation system
- evaluation criteria
- image quality
- image compression
- web search
- multimedia
- information retrieval