A Random Forest decision-support tool for managing education that uses CSV-based LMS interaction data as its basis

Abstract

This study presents a decision-support application that employs a machine learning model based on data regarding how engaged students are to suggest individualized treatments. The system takes in CSV files that contain information about participation rates, how often students use the LMS, how many assignments they do, and feedback from teachers. The model makes predictions about what new students should do based on labeled examples. For example, it might say "Mentorship recommended" or "Alert—Immediate action needed." The program has a user-friendly GUI made with PySimpleGUI that makes it easy to submit files, train the model, and show predictions. You can make up the data you use for training, but the format works with exports from popular LMS platforms like Google Classroom and Moodle. This method gives educational management an easy way to use predictive analytics to keep an eye on student involvement. The system helps data-driven educational administration and change efforts by automatically sorting students and showing early warning signs.