Resumen de Crowdsourcing Translation Technology Based on Recurrent Neural Network Model with Bilingual Constraints

Hui Zhao, Ali Bin Selama, Kaiyisah Hanis Binti Mohd Azm

• Neural network algorithm is a convenient and fast computer intelligent algorithm, and it is widely used in various fields of science and technology with the development of computer technology. In order to study Crowdsourcing Translation, a recurrent neural network model based on bilingual constraints was proposed, and this model was applied to the decoding process of Crowdsourcing Translation; then, a step-by-step and semi-supervised training method was proposed to train the established model; in addition, the expressions of phrases in translation are explored, and an expression based on translation confidence was proposed; finally, the performance of the whole translation was verified by bilingual constraints from Chinese to English, and the validity and correctness of the method were proved by the analysis of the experimental results.