A Novel Deep Learning Method for Obtaining Bilingual Corpus from Multilingual Website

Table 3

Statistics of the size and precision of parallel sentences extracted from multilingual websites.

Model	Training corpus	#sentences	#precision
Bitextor&LSTM	30,000	117,900	0.70
	40,000	124,200	0.70
Ours&LSTM	30,000	120,200	0.81
	40,000	127,900	0.82