The Investigation of Student's Online Learning Adaptability Level Prediction Based on AutoML

In the rapidly evolving realm of online education, developing an accurate method for assessing students' adaptability is of paramount importance to the educational field. It enables educational institutions to better understand student performance and needs, allowing for the provision of timely support and resources. Therefore, this study employed Automated Machine Learning (AutoML) technology to predict students' adaptability to online learning. Utilizing Baidu's AI development platform named EasyDL, this research conducted an in-depth analysis of 1, 205 data samples provided by the Kaggle platform. The model's performance was evaluated using key indicators e.g. accuracy, precision, recall, and F1-score, and it was compared with traditional machine learning methods to establish its superiority. The experimental results demonstrate that AutoML technology effectively predicts students' adaptability to online learning, with the model achieving an accuracy rate of 90.9%. While there is still room for improvement in predicting high adaptability levels in students, the model significantly underscored the importance of critical features such as age, location, type of educational institution, and class duration. These findings not only validate the efficacy of AutoML technology in the educational sector but also lay an empirical foundation for future research.