With the rapid development of information technology, the field of education is gradually adopting intelligent technology to enhance the teaching effect. This study proposes a multi-scenario teaching strategy optimization path based on reinforcement learning for English in the higher vocational railroad industry, which recommends personalized learning paths for learners by combining the Deep Knowledge Tracking (DKT) model with the Reinforcement Learning (RL) method. In the experiments, the model is validated on different datasets, in which the AUC in the Skill-builder-data-2023-2024 dataset reaches 0.83837, and the accuracy is 0.75654. By comparing with the traditional models (e.g., BKT, DKT, KNN, etc.), the method of this paper demonstrates a significant advantage in the accuracy of the learning path recommendation. At the same time, it is also able to adjust the recommendation strategy according to the dynamic performance of the learner to further optimize the learning effect. The results show that the learning path recommendation based on reinforcement learning can significantly improve the degree of personalization and adaptability of higher vocational English teaching, and has high practical value.
