作者: Jinying Kong , Yating Yang , Xi Zhou , Lei Wang , Xiao Li
DOI: 10.1007/978-3-319-50496-4_12
关键词:
摘要: The problem of rare and unknown words is an important issue in Uyghur-Chinese machine translation, especially using neural translation model. We propose a novel way to deal with the words. Based on pointers over input sequence, our approach which consists preprocess post-process can be used all Pre-process modify corpus extend ability pointer network, post- process retranslating raw by phrase-based model or wordlist. Experiment show that proposed this paper get higher BLEU score than MT.