A Comparison of English Learners’ Translation Bias Using Neurosemantic Analysis

Abstract

This paper first explores translation bias and bias analysis and comes up with the characteristics of English translation bias. Secondly, the neural semantic analysis method is created by combining corpus technology and a neural machine translation model, and translation optimization is done based on neural semantic analysis. Finally, the performance of the neural semantic analysis method and the translation effect using the neural semantic analysis method are experimented with, and a comparative analysis of bias is carried out. The results show that the average accuracy of document recognition by the method based on neural semantic analysis is around 0.92, the difference between different languages by using neural semantic analysis is around 0.01, and the bias degree of translation techniques and translation units using the neural semantic method has been reduced by 0.17~0.24, and 0.10~0.17, respectively. The use of neural semantic analysis is able to reduce the translation bias of English language learners and thus improve the accuracy of translation.

Keywords

• corpus technology
• neural semantic analysis
• neural machine translation
• translation optimization
• translation bias
• 01a75