Bilingual Evaluation Understudy (BLEU) Performance Measure
A Bilingual Evaluation Understudy (BLEU) Performance Measure is an MT Metric for evaluating machine translation systems that is based on a modified unigram's precision.
- AKA: BLEU Score, BLEU.
- Context:
- It was initially developed by (Papineni et al., 2002).
- It can (typically) not take Intelligibility or Grammatical Correctness into account.
- It can be defined as the unigram's precision $P = \dfrac{m}{w_{t}}$ where $m$ is the number of words in both candidate and reference translations, and $w_r$ is the number of words in the candidate translation alone.
- Example(s):
- Counter-Example(s):
- See: Self-BLEU Measure, Text Corpus, Machine Translation, Correlation, Machine Summarization.
References
2018
- (Wikipedia, 2018) ⇒ https://en.wikipedia.org/wiki/BLEU Retrieved:2018-8-27.
- BLEU (bilingual evaluation understudy) is an algorithm for evaluating the quality of text which has been machine-translated from one natural language to another. Quality is considered to be the correspondence between a machine's output and that of a human: "the closer a machine translation is to a professional human translation, the better it is" – this is the central idea behind BLEU. BLEU was one of the first metrics to claim a high correlation with human judgements of quality, and remains one of the most popular automated and inexpensive metrics.
Scores are calculated for individual translated segments — generally sentences — by comparing them with a set of good quality reference translations. Those scores are then averaged over the whole corpus to reach an estimate of the translation's overall quality. Intelligibility or grammatical correctness are not taken into account.
BLEU’s output is always a number between 0 and 1. This value indicates how similar the candidate text is to the reference texts, with values closer to 1 representing more similar texts. Few human translations will attain a score of 1, since this would indicate that the candidate is identical to one of the reference translations. For this reason, it is not necessary to attain a score of 1. Because there are more opportunities to match, adding additional reference translations will increase the BLEU score.
- BLEU (bilingual evaluation understudy) is an algorithm for evaluating the quality of text which has been machine-translated from one natural language to another. Quality is considered to be the correspondence between a machine's output and that of a human: "the closer a machine translation is to a professional human translation, the better it is" – this is the central idea behind BLEU. BLEU was one of the first metrics to claim a high correlation with human judgements of quality, and remains one of the most popular automated and inexpensive metrics.
2017
- (Manning & Socher, 2017k) ⇒ Christopher Manning, and Richard Socher. (2017). “Lecture 11 - Further Topics in Neural Machine Translation and Recurrent Models.”
2011
- (Madnani, 2011) ⇒ Nitin Madnani. (2011). “iBLEU: Interactively Debugging and Scoring Statistical Machine Translation Systems.” In: Proceedings of the 5th IEEE International Conference on Semantic Computing (ICSC 2011). DOI: 10.1109/ICSC.2011.36.
2002
- (Papineni et al., 2002) ⇒ Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. (2002). “Bleu: A Method for Automatic Evaluation of Machine Translation.” In: Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics (ACL 2002). DOI:10.3115/1073083.1073135.