Neural machine translation method based on similarity perception

Abstract

The invention provides a neural machine translation method based on similarity perception, and belongs to the technical field of natural language processing machine translation. The method comprises the following steps of firstly, creating a structure translation memory bank and a corresponding structure similarity algorithm; constructing a template translation memory bank and a corresponding template similarity algorithm; and then pre-identifying high-potential statements of character strings, structures and template dimensions in the test set; constructing multi-dimensional similarity prioriknowledge, and performing multi-dimensional similarity retrieval on all parallel statements in the training set. A posteriori regular objective function is utilized, discrete similarity priori knowledge is fused into a neural machine translation objective function, parameters of the priori knowledge are continuously iterated and updated, and a training process is guided. Finally, the multi-dimensional high-potential to-be-translated statements are translated by utilizing the trained neural translation model. The most similar sentences can be obtained in a finer-grained manner, and the reviewtime of human translators is shortened.

Description

technical field [0001] The present invention relates to a technique for modeling the multidimensional similarity of parallel corpora in neural machine translation and can identify high-potential sentences in a test set and optimize their corresponding translation performance, in particular to a neural machine translation system based on similarity perception The invention and method belong to the technical field of natural language processing machine translation. Background technique [0002] At present, neural machine translation is widely used in multi-scenario computer-aided translation tasks because it is better than traditional statistical machine translation in a variety of natural languages. However, most existing neural machine translation methods focus on how to improve the overall translation performance, paying little attention to the workload of human translators. [0003] In computer-assisted translation scenarios, human translators receive translations generat...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to solve the technical problem that the existing machine translation system cannot identify and optimize high-potential sentences for sentences to be translated in practical applications, resulting in redundant review and heavy post-translation editing workload for human translators. A neural machine translation method based on similarity perception

Images