A Study on the Translation of English Network Terms in the Age of Social Media

Abstract

In the era of social media, the Internet has given rise to network language, and English network terms have gradually become popular, which are not only easy to understand but also rich in interest and highly favored by the majority of Internet users. In this paper, after systematically integrating the translation features and translation strategies of English network terms in the social media era, we build a social network terminology database based on NILE automatic recognition technology and propose a machine translation method based on the features of English network terms and migration learning. The translation results show that the model of this paper has the highest average BLEU value of 16.02% for the translation of the target network terminology, and the model of this paper has the least frequency of BLEU value of 0% for each translation, which is 2 times, and it has the best ability of “bottoming out”. In addition, the model can skillfully transform verbs and pronouns and is good at translating special words such as superlatives so as to make the translations closer to the expression habits of the target language and meet the social needs of the English Internet language.

Keywords

• automatic nile recognition
• transfer learning
• machine translation
• english online language features
• 68m15