Authors: Associate Professor Shilpa K
Abstract: Purpose: Student performance is a key factor in addressing educational issues and evaluating learning outcomes. Educational Data Mining (EDM) has emerged to leverage data knowledge for enhancing educational systems. Objectives: This study aims to improve the predictive accuracy of academic success using eight popular machine learning techniques. It focuses on developing methods for analyzing data from learning environments to enhance student experiences and provide precise insights. Methodology/Design/Approach: The study utilizes EDM to analyze data from platforms like Learning Management Systems (LMS), Massive Open Online Courses (MOOC), and Intelligent Tutoring Systems (ITS). Eight machine learning techniques, including k-nearest neighbors (KNN), support vector machine (SVM), and random forest (RF), are employed. Performance indicators such as recall, accuracy, precision, and F-measure evaluate the prediction model's effectiveness. Findings/Result: EDM aids in automated grading, recommendations, and adaptive systems, significantly improving student performance. Predicting students' grades helps identify at-risk individuals, allowing timely interventions that benefit parents, educators, and students. Originality/Value: This study underscores the significance of EDM in extracting meaningful insights from academic data and improving educational outcomes through advanced predictive models.
