Meteor, M-BLEU and M-TER: Evaluation Metrics for High-Correlation with Human Rankings of Machine Translation Output.
Abhaya AgarwalAlon LaviePublished in: WMT@ACL (2008)
Keyphrases
- machine translation
- evaluation metrics
- high correlation
- lexical cohesion
- average precision
- precision and recall
- highly correlated
- cross lingual
- natural language processing
- chinese english
- text summarization
- information extraction
- target language
- cross language information retrieval
- statistical machine translation
- machine translation system
- natural language
- word alignment
- human judgments
- learning to rank
- mt evaluation
- query translation
- evaluation measures
- parallel corpora
- relevance judgments
- source language
- named entity recognition
- digital libraries