Article: AI-driven English translation: leveraging machine learning and deep learning for enhanced accuracy Journal: International Journal of Information and Communication Technology (IJICT) 2025 Vol.26 No.18 pp.1 - 17 Abstract: The quick growth of AI has greatly affected the field of machine translation, which in turn has resulted in more accurate and context-aware English translations. This article suggests an English translation framework that is based on the combination of machine learning (ML) and deep learning (DL). The study presents a variety of neural architectures including transformer-based models (e.g., BERT, GPT) and neural machine translation (NMT) systems by dealing with issues about translation fluency and contextual understanding. The authors use reinforcement learning (RL) and fine-tuning that are two of the machineries in their laboratory to bolster translation in the case of low-resource languages and technical writing. The suggested hybrid model leverages the power of both rule-based linguistic processing and AI technology for error avoidance and added real-time translation performance. As observed from the experimental results, the new model definitely has the edge as compared to the traditional statistical and rule-based systems. It gives out the highest BLEU and METEOR scores. This study is truly a solid basis for the way forward towards fully AI-driven multilingual translation systems. Inderscience Publishers - linking academia, business and industry through research

Title: AI-driven English translation: leveraging machine learning and deep learning for enhanced accuracy

Authors: Qing Zhao

Addresses: English Language and Literature, Xi'an Fanyi University, Xi'an, 710105, China

Abstract: The quick growth of AI has greatly affected the field of machine translation, which in turn has resulted in more accurate and context-aware English translations. This article suggests an English translation framework that is based on the combination of machine learning (ML) and deep learning (DL). The study presents a variety of neural architectures including transformer-based models (e.g., BERT, GPT) and neural machine translation (NMT) systems by dealing with issues about translation fluency and contextual understanding. The authors use reinforcement learning (RL) and fine-tuning that are two of the machineries in their laboratory to bolster translation in the case of low-resource languages and technical writing. The suggested hybrid model leverages the power of both rule-based linguistic processing and AI technology for error avoidance and added real-time translation performance. As observed from the experimental results, the new model definitely has the edge as compared to the traditional statistical and rule-based systems. It gives out the highest BLEU and METEOR scores. This study is truly a solid basis for the way forward towards fully AI-driven multilingual translation systems.

Keywords: AI-driven translation; machine learning; ML; deep learning; DL; neural machine translation; NMT; reinforcement learning; RL; contextual adaptation; computational linguistics.

DOI: 10.1504/IJICT.2025.146691

International Journal of Information and Communication Technology, 2025 Vol.26 No.18, pp.1 - 17

Received: 17 Mar 2025
Accepted: 07 Apr 2025
Published online: 13 Jun 2025 *

Title: AI-driven English translation: leveraging machine learning and deep learning for enhanced accuracy

Keep up-to-date