A Predictive Model of Learning Effectiveness in Flipped Classroom Mode: An Exploration of Higher Vocational English Learning Based on Machine Learning

Abstract

Abstract Taking English microclasses as an example, this paper analyzes the practical operation of flipped classroom teaching in the reform of higher vocational English teaching from the three phases of pre-course, in-course and post-course. Comparing and analyzing the advantages of each fusion algorithm, the Stacking model fusion algorithm is selected to construct a multi-model fusion prediction model of students’ learning effectiveness, and the experimental process of students’ learning effectiveness prediction model based on Stacking fusion is summarized. The algorithmic performance of each machine learning prediction model is determined using each evaluation index. The multi-model fusion learning effectiveness prediction model is employed to predict and analyze the overall and individual effectiveness of English learning by organizing students’ English learning data. Combined with the prediction results of the flipped classroom platform data, the overall performance of the multi-model fusion prediction model is more stable, with a more balanced distribution in the range of 0.7~0.9, which can obtain better accuracy performance than LR, GBDT and XGBoost, and is more capable of predicting the students’ learning effectiveness in terms of the stages of learning (certified, grade, and total_time) in real life. Prediction.