Practical Application and Effect Evaluation of an AI-Empowered Training Model for Enhancing Language and Writing Abilities

Abstract

Constant interaction, feedback-based correction and task matching are essential for improving language and writing ability. The existing training models have problems such as poor refinement of feedback, fixed learning paths and difficulty providing personalized content. Artificial intelligence provides technical help to reform the language and writing training, but merely stacking tools cannot upgrade the training model. In this paper, we construct an AI-enabled training model for improving language and writing ability according to cognitive linguistics and adaptive learning theory from three aspects: theoretical foundation; operating mechanism; and effect evaluation and iterative optimization. We clarify the cognitive characteristics of learners' language processing and constraints of technological intervention, analyse the implementation path of multimodal corpus supply, adaptive learning paths, immediate feedback and human–machine collaboration, establish a multidimensional evaluation and iterative closed loop, choose 42 higher vocational students who majored in finance and business to carry out an 8-week small-sample pilot practice. Our research shows that our model has achieved internal unity among objectives, tasks, content, feedback and learning paths, however, there are some drawbacks (e.g., insufficient AI evaluation of pragmatic appropriateness in workplace scenarios), which can provide a model reference for AI-enabled language and writing training in vocational colleges without facing the problem of technological idling or content generalization.