E-learning is an interactive online learning mode that can enhance students' interest in learning in an entertainment environment. This article aims to develop an English vocabulary learning recommendation system based on decision tree algorithm and naive Bayesian algorithm to provide personalized learning suggestions and help learners learn and remember words more effectively. Firstly, a large amount of English vocabulary data was collected, preprocessed, and feature extracted. Decision tree algorithms were selected as the foundation, and through analysis of existing data, a decision tree was generated to classify new data. Using learners' characteristics (such as age, learning objectives, etc.) and learning history (such as learned words, learning time, etc.) as inputs, a personalized recommendation model was constructed, This model can recommend suitable learning content for learners based on their personalized needs and learning situation. In order to better understand learners' learning progress, a naive Bayesian model was trained using learners' learning history and progress information to analyze their current learning progress and predict their future learning situation. Through testing and evaluation of actual learners, the recommendation system performs well in providing personalized learning suggestions, and has a significant improvement effect on learners' vocabulary learning.
Read full abstract