Design Of Translation Ambiguity Elimination Method Based On Recurrent Neural Networks

The ambiguity of language inevitably leads to the ambiguity of translation, and how to deal with translation ambiguity has become a persistent focus of attention for both human translation and machine translation. Traditional machine translation mainly adjusts the threshold of ambiguity similarity to deal with translation ambiguity, but the effect is not ideal. The machine translation model based on recurrent neural networks provides us with a new perspective. In this new perspective, the candidate set calculates the similarity, obtains the source language and target language of the reference translation, and then nested in the neural network to complete the ambiguity elimination in language translation. This translation model based on recurrent neural networks effectively eliminates the gradient imbalance problem generated during the translation ambiguity process. Comparative experimental results also show that with a reasonable setting of the similarity threshold, the advantages of the new method are more evident and can better improve the translation results.