Improvement of English-Chinese bilingual learning by integrating semantic analysis and neural machine translation

Abstract

Abstract To improve the effectiveness of English-Chinese bilingual learning, this study introduces an optimization model that combines semantic analysis with neural machine translation (NMT). A series of comprehensive experiments were conducted to evaluate its performance. The results demonstrated that the proposed model outperformed existing approaches across multiple key metrics, including reading comprehension, translation accuracy, and learning efficiency. Specifically, the optimized model achieved a reading comprehension score of 85.0, a grammar accuracy rate of 82.5%, and a vocabulary score of 80.1. For translation performance, it attained a sentence-level accuracy of 84.3%, paragraph coherence of 81.5%, and a cultural understanding score of 78.9. In terms of learning efficiency, the model reduced average learning time to 45.3 min, reached a knowledge retention rate of 80.2%, and improved autonomous learning ability to 78.5%. Learner satisfaction also improved, with an interest level score of 4.3, a teaching method satisfaction score of 4.0, and an overall satisfaction score of 4.2 (on a 5-point scale). Compared with rule-based translation systems and models using word embeddings or basic neural networks, the proposed model demonstrated clear advantages in all areas of evaluation. Overall, this study offers a valuable contribution to the advancement of computer-assisted language learning technologies.

Keywords

• Fusion semantic analysis
• Neural machine translation
• English-Chinese bilingual learning
• Language comprehension ability