Academic Supervision and Risk Assessment Based on Moodle LMS Data

With the application of Learning Management Systems(LMS) in online learning platforms, there is already a large amount of data to record student operations. Many researchers have done some research based on this type of data and obtained some valid predictors. First, we analyze the theoretical feasibility of this research and combine other established research results to determine our research objectives. Then our data comes from a learning platform based on Moodle LMS in Chongqing. The main research objects include 8 blended courses and 638 students. After some experiments, we established a general multiple linear regression model with interaction effects to analyze the relationship between predictor and final grade. After some experiments, we establish a ordinary multiple linear regression model with interaction effects to analyze the relationship between predictors and final grades. Our model has a good track record in predicting student achievement. Therefore, we hope to establish a system of academic supervision and reminders on the basis of this model to help students better manage their own learning and strengthen the supervision of teachers. Since the data may vary greatly at different levels of education, we need to improve the model in the future to achieve greater portability and support more data information.