Application of Artificial Intelligence in Computer-Assisted English Vocabulary Translation

Abstract

In the era of artificial intelligence, machine translation based on neural networks and deep learning is an important tool for computer-aided English translation. However, the accuracy of existing word-based domain feature learning methods in domain recognition is low, which reduces the efficiency and accuracy of English word translation. As a result, this paper proposes a multi-domain neural machine translation method based on word domain feature sensitivity to address the problem of translation models considering word domain features in isolation when there is no apparent domain tendency in a sentence and low domain discrimination accuracy. Compared with other models, the proposed model can extract contextual features on top of domain features of words and calculate enhanced domain proportions for each word to guide translation generation. A significant improvement in word translation accuracy is also observed in the proposed model compared with the baseline model, as well as a stronger learning ability, which has significant potential for use in English word translation.