This paper constructs an intelligent English translation teaching model based on a multi-strategy bee colony algorithm, capable of realising personalised teaching by dynamically adjusting content and learning paths.Experimental data indicate that the model significantly enhances student performance; average scores in large classes increased by over 13 points (approximately 19.6%), demonstrating strong scalability.Compared to traditional algorithms like PSO and GAE, the model achieves stability within just 18 iterations, significantly optimising error rates compared to previous fluctuations.Furthermore, it drastically reduces task completion timehandling 60-word tasks in under one hour, whereas traditional neural models require over nine.While senior students exhibit rapid short-term gains and juniors show stable long-term improvement, the model ultimately validates itself as a highly efficient, precise, and personalised solution for modernising translation teaching.
Wang et al. (Thu,) studied this question.