Utilizing Machine Learning Models to Predict Student Performance From LMS Activity Logs

Abstract

In recent years, the application of data mining techniques on educational data has grown in importance. Educational data mining can be used to find hidden patterns in students’ academic conduct and predict future success by examining previous data. Because more technical tools are being used to enhance the learning environment, including learning management systems (LMS), the importance of educational data mining is growing for educational institutions. The purpose of this study is to employ data mining techniques to analyse pupil behaviour patterns and predict how well they would perform academically. According to the findings of this study, there is a considerable correlation between student performance and a number of different factors, such as resource (page) views, activity gaps, grades from the previous semester, grades from prerequisite courses, and evaluations of first-term tests. Teachers and educators can use this study to spot students who need extra assistance so they can intervene.

Keywords

• Activity logs
• data mining
• sentiment analysis
• machine learning
• neural networks
• emotion recognition