Reforming Traditional Medicine Translation Courses in Higher Education through a Human–Machine Collaborative Blended Teaching Model: An Empirical Study

Abstract

The need for qualified translators with expertise in traditional medicine has increased due to China's fast globalisation. This study investigated a human–machine collaborative blended teaching model using the Feiyi platform to improve undergraduate students' translation performance in light of the drawbacks of conventional translation techniques. The model, which was based on constructivist and social learning theories, combined guided machine translation exercises, in-person training, and online terminology databases. The model's effect on translation efficiency and accuracy was evaluated using a quasi-experimental design. Results from tests conducted before and after the intervention (N = 25) showed notable gains in efficiency (pre-test M = 43.0 min, post-test M = 31.0 min, p <.001) and overall translation accuracy (pre-test M = 68.2%, post-test M = 76.5%, p <.001). Significant improvements in cultural appropriateness (d = 1.51) and terminological accuracy (d=1.57) were noted, suggesting the model successfully addressed important issues in translating traditional medicine. These results show that the blended model promotes positive affective outcomes and enhances quantifiable translation performance. Among the ramifications are the methodical integration of machine-assisted phases to standardise specialised language and encourage the development of independent, critical translation abilities.