Fuzzy Decision Tree Optimization for Personalized Hybrid English Teaching in Smart Classrooms

It is difficult for hybrid English teaching to target individual differences among students. Based on the optimization method of fuzzy decision tree, this paper uses its fuzzy decision-making ability to dynamically adjust teaching strategies to achieve flexible and personalized teaching. Learning behavior data are collected through the smart classroom platform. In the fuzzy logic application, the learning behavior data are converted into fuzzy information, and triangular and trapezoidal membership functions are utilized to classify and fuzzify it. The construction of the decision tree is based on the information gain ratio, the best split point is selected, and the model structure is optimized through cost complexity pruning to prevent overfitting. For complex student learning behaviors, the teaching strategy is dynamically adjusted in combination with the hierarchical decision-making ability of fuzzy decision trees. Finally, Hadoop and Spark architectures are combined for dynamic tracking to achieve real-time learning progress monitoring and personalized feedback. The research results show that the system takes 0.45[Formula: see text]s to infer 2,000 inputs; the average student score increases by 9.01%; the learning time increases by 15%; the participation increases by 15%. This method can significantly enhance students’ learning effects and personalized learning experience, and provide an effective solution for the optimization of hybrid teaching in English smart classrooms.