Meaningless yet meaningful: Morphology grounded subword-level NMT
Meaningless yet meaningful: Morphology grounded subword-level NMT
复制标题
无意义却有意义：基于形态学的子词级 NMT
DOI：
--
复制
发表时间：
2018
期刊：
影响因子：
--
通讯作者：
P. Bhattacharyya
中科院分区：
文献类型：
--
作者：
Tamali Banerjee;P. Bhattacharyya
我们探索使用两个独立的子系统字节对编码（BPE）和Morfessor作为基本单位的子词级神经机器翻译（NMT）。我们发现，对于语言上遥远的语言对基于Morfessor的分割算法产生的翻译质量显着优于BPE。然而，对于接近的语言对，基于BPE的子词NMT可能比基于Morfessor的子词NMT翻译得更好。我们提出了这两种分割算法Morfessor-BPE（M-BPE）的组合方法，它在BLEU评分方面优于这两个基线系统。我们的研究结果得到了三种语言对：英语-印地语，孟加拉语-印地语和英语-孟加拉语实验的支持。
We explore the use of two independent subsystems Byte Pair Encoding (BPE) and Morfessor as basic units for subword-level neural machine translation (NMT). We show that, for linguistically distant language-pairs Morfessor-based segmentation algorithm produces significantly better quality translation than BPE. However, for close language-pairs BPE-based subword-NMT may translate better than Morfessor-based subword-NMT. We propose a combined approach of these two segmentation algorithms Morfessor-BPE (M-BPE) which outperforms these two baseline systems in terms of BLEU score. Our results are supported by experiments on three language-pairs: English-Hindi, Bengali-Hindi and English-Bengali.