Tackling Sparse Data Issue in Machine Translation Evaluation

We illustrate and explain problems of n-grams-based machine translation (MT) metrics (e.g. BLEU) when applied to morphologically rich languages such as Czech. A novel metric SemPOS based on the deep-syntactic representation of the sentence tackles the issue and retains the performance for translation to English as well.

• Type: Journal
• Year: 2010
• Where: ACL
• Authors: Ondrej Bojar, Kamil Kos, David Marecek
• Keywords: ACL 2010, Computational Linguistics, Deep-syntactic Representation, N-grams-based Machine Translation, Rich Languages