Method of Augmenting Korean Classical Literature Corpus for Machine Translation Model

摘要

The present invention relates to a method for augmenting a Chinese text book corpus for a machine translation model, and more particularly, using a parallel corpus constructed for learning at least one of gimbal points, Chinese character noise, translation stage noise, reverse translation, sentence segmentation, and pre-extraction techniques. It relates to a method of augmenting a corpus using a technique. The Chinese text book corpus augmentation method for a machine translation model according to an embodiment of the present invention comprises a parallel corpus built for learning that is a starting word (source) of an input unit, In the augmentation part, any one or more techniques of punctuation marks (punctuation marks), Chinese characters (original characters) noise (A), translation stage noise (B), reverse translation (C), sentence division (D), and dictionary extraction (E) augmented, By outputting the target language (target) to the output unit, Characterized in increasing the amount of corpus.