METHOD AND APPARATUS FOR MACHINE TRANSLATION USING NEURAL NETWORK AND METHOD FOR LEARNING THE APPARTUS

摘要

Disclosed are a machine translation method and a machine translation apparatus using a neural network model. The machine translation apparatus extracts keyword-related information from the source sentence, and searches for additional sentences related to the source sentence based on the extracted keyword-related information. The machine translation apparatus obtains a first vector value and a second vector value from a source sentence and an additional sentence, respectively, by using encoders based on a neural network model. The machine translation apparatus outputs a target sentence corresponding to a translation result of a source sentence based on a first vector value or a second vector value using a decoder based on a neural network model.