Abstract | ||
---|---|---|
N-gram-based metrics have been used widely in automatic evaluation of machine translation. However, most of them also lose merits due to the strict policy of matching of n-grams. Especially, the policy of exact matching leads to take synonyms as totally different words and thus give unreasonable estimation. This paper introduces fuzzy matching for n-grams, which refers to a semantic similarity function based on WordNet. And it is used to find a match with the highest similarity when incorporated into BLEU, the representative of n-gram-based evaluation metrics. Since WordNet can contribute more to high-order n-grams and fuzzy matching can perform well even with fewer references, experiments on MTC Part 2 (LDC2003T17) show our proposed method can greatly improve correlation between BLEU and human evaluation both at segment-level and document-level. Furthermore, BLEU incorporating fuzzy matching achieves more significant improvement at document-level evaluation. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1007/978-3-642-36337-5_5 | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
Keywords | Field | DocType |
semantic similarity,n-gram-based metrics,human evaluation,automatic evaluation,document-level evaluation,n-gram-based mt evaluation,highest similarity,fuzzy matching,strict policy,n-gram-based evaluation metrics,exact matching,machine translation,wordnet | Semantic similarity,BLEU,Pattern recognition,Evaluation of machine translation,Computer science,Machine translation,Approximate string matching,n-gram,Artificial intelligence,WordNet | Conference |
Volume | Issue | ISSN |
7717 LNAI | null | 16113349 |
Citations | PageRank | References |
0 | 0.34 | 11 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Liangyou Li | 1 | 3 | 2.74 |
Zhengxian Gong | 2 | 69 | 8.49 |