The use of deep learning and AIoT technology in loan word translation. (5th March 2022)

• Record Type: Journal Article
• Authors: Xu, Xiaoyan; Yang, Zhenhuan
• Is Part Of: International journal of grid and utility computing. Volume 13:Number 1(2022)
• Page Start: 49
• Page End: 56
• Publication Date: 2022-03-05
• Subjects: intelligent translation -- loan word translation -- deep learning -- intelligent internet of things; Electronic data processing -- Distributed processing -- Periodicals; Electronic commerce -- Management -- Computer programs -- Periodicals; 004.605
• Languages: English
• ISSNs: 1741-847X

Abstract

The purpose is to translate loan words accurately, standardise their use, and better integrate with Chinese culture. The role of intelligent translation in loan words translation is studied based on deep learning and intelligent Internet of things. Moreover, a new neural translation mechanism of divide and conquer strategy is proposed based on the original Artificial Intelligence (AI) translation technology. Its translation quality is tested and compared with manual translation. The new neural translation mechanism can achieve the accuracy of human daily language. The accuracy rate of system Mobile Number Portability (MNP) identification is 74.67%, the recall rate is 71.42% and the F-value is 74.01%. Moreover, the average time is 0.01s/character. Therefore, the research results suggest that the constructed AI translation can more efficiently complete the translation task, save a lot of time cost and labour cost and produce a crucial reference for the intelligent development of the translation industry.