Chinese-English Bidirectional Semantic Understanding and Translation Method Based on Convolutional Neural Network

With the advancement of globalization, the demand for two-way translation between Chinese and English is increasing. Existing machinetranslation technologies have problems in semantic understanding, accuracy and fluency of translation. To this end, this paper proposes a Chinese-English two-way semantic understanding and translation method based on convolutional neural network (CNN). This paper first uses a deep convolutional neural network to extract features and semantically represent text. Then, the sequence-to-sequence model is improved by using the bidirectional encoder-decoder structure. Finally, this paper combines the attention mechanism and adaptive learning strategy to improve the translation quality. From the experimental conclusions, the score of the CNN model gradually increased to 0.67 in the BLEU (Bilingual Evaluation Understudy) score. In terms of the METEOR score, the score of the CNN model rose to 0.79. In the above data conclusions, the effectiveness of the CNN-based translation method in improving translation quality and semantic understanding shows its great application potential in the field of machine translation.