Analysis of Semantic Deviation in AI Translation and Linguistic Optimization Paths
DOI:
https://doi.org/10.71451/ISTAER2522Keywords:
AI Translation; Semantic Bias; Context Understanding; Training Data; Semantic ReasoningAbstract
With the rapid development of artificial intelligence technology, AI translation plays an increasingly important role in cross-language communication. However, AI translation still faces the problem of semantic bias, which affects the accuracy and fluency of translation. This paper analyzes the main manifestations and causes of semantic bias in AI translation, explores the limitations of the model itself, the impact of training data problems and language structure differences on translation, and proposes corresponding language optimization paths, including improving context understanding ability, optimizing training data and models, enhancing semantic reasoning ability, and strengthening cultural adaptation and localization. Studies have shown that through multi-dimensional optimization, semantic bias can be effectively reduced and translation quality can be improved. Finally, the future development direction of AI translation technology is prospected, especially the further optimization path in a multilingual and multicultural environment, which provides new ideas and directions for future AI translation research.
References
[1] Cao, L., & Fu, J. (2023). Improving efficiency and accuracy in English translation learning: Investigating a semantic analysis correction algorithm. Applied Artificial Intelligence, 37(1), 2219945. DOI: https://doi.org/10.1080/08839514.2023.2219945
[2] Mohamed, Y. A., Khanan, A., Bashir, M., Mohamed, A. H. H., Adiel, M. A., & Elsadig, M. A. (2024). The impact of artificial intelligence on language translation: a review. Ieee Access, 12, 25553-25579. DOI: https://doi.org/10.1109/ACCESS.2024.3366802
[3] Zhang, B., & Liu, Y. (2022). Construction of English translation model based on neural network fuzzy semantic optimal control. Computational Intelligence and Neuroscience, 2022(1), 9308236. DOI: https://doi.org/10.1155/2022/9308236
[4] Cao, J., Li, M., Li, Y., Wen, M., Cheung, S. C., & Chen, H. (2022). SemMT: a semantic-based testing approach for machine translation systems. ACM Transactions on Software Engineering and Methodology (TOSEM), 31(2), 1-36. DOI: https://doi.org/10.1145/3490488
[5] Huang, D., & Xiang, S. (2024). Multi-objective optimal routing algorithm translation conversion theory subtitle translation research. Journal of Computational Methods in Sciences and Engineering, 24(6), 3892-3901. DOI: https://doi.org/10.1177/14727978241299638
[6] Chen, Y. (2023). Analyzing the design of intelligent English translation and teaching model in colleges using data mining. Soft Computing, 27(19), 14497-14513. DOI: https://doi.org/10.1007/s00500-023-09096-7
[7] Stanovsky, G., & Dagan, I. (2018). Semantics as a foreign language. In Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing (pp. 2412-2421). DOI: https://doi.org/10.18653/v1/D18-1263
[8] Shujie, Z. H. A. O. (2025). Applications and Challenges of ChatGPT in Translation: A Multidimensional Exploration of Theory and Practice. Sino-US English Teaching, 22(2), 35-41. DOI: https://doi.org/10.17265/1539-8072/2025.02.001
[9] Zheng, H., Xu, K., Zhou, H., Wang, Y., & Su, G. (2024). Medication recommendation system based on natural language processing for patient emotion analysis. Academic Journal of Science and Technology, 10(1), 62-68. DOI: https://doi.org/10.54097/v160aa61
[10] Chandra, R., & Kulkarni, V. (2022). Semantic and sentiment analysis of selected Bhagavad Gita translations using BERT-based language framework. IEEE Access, 10, 21291-21315. DOI: https://doi.org/10.1109/ACCESS.2022.3152266
Downloads
Published
Issue
Section
License
Copyright (c) 2025 International Scientific Technical and Economic Research
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
This work is licensed under the Creative Commons Attribution International License (CC BY 4.0).
