Machine Automatic Translation Evaluation Model based on Big Data Algorithm

The development of various fields such as artificial intelligence (AI) and machine translation (MT)is accelerating, with people processing information in different languages to make up for the shortcomings of human translation in speed and cost. In the process of undirected learning, the denoising autoencoder will be used to learn bilingual words synchronously. In this paper, unsupervised and supervised learning methods are used to calculate the weight and bias of each level, respectively, and establish the regression model of translation features. In order to evaluate the quality of translation, a machine automatic translation(MAT) evaluation system is designed. The research results show that in the comparison of the translation time before and after using the system for different times of testing, the average translation time before using the system is 45.58 minutes. This suggests that the machine automatic translation evaluation model based on big data algorithms can help improve the translation speed.