Design of an Automatic Translator Based on Graph Convolutional Neural Network and Backtranslation Process
In order to solve the problem of fast translation between Chinese and French,a graph convolutional neural network combined with backtracking process is proposed for automatic translation.The attention mechanism will be introduced in the process,and backtracking methods will be used to expand the data of the translator.The experimental results show that the average Rouge-L score of the research method in French Chinese translation tasks remains above 0.908 under different vocabulary counts;When con-ducting translation time testing,the research method took 178ms when the Chinese translation method reached 60 words.The results indicate that the automatic translator designed by the research institute has good translation accuracy in both directions when transla-ting text between Chinese and French,and can complete translation tasks in a shorter time.The translation results are more in line with human actual reading logic,and can provide effective technical references for translation tasks.
万方数据
维普
