PREDICTIVE MODEL FOR STUDENT DROPOUT RISK IN ONLINE LEARNING ENVIRONMENT

Abstract

Abstract
Online learning environments have become an important part of modern education by providing flexible access to courses, learning materials, assessments, and communication facilities. However, student dropout remains a major challenge because learners may discontinue their courses due to poor academic performance, low engagement, irregular participation, lack of interaction, personal difficulties, or technical problems. This paper proposes a predictive model for identifying student dropout risk in online learning environments using machine learning techniques. The proposed approach analyzes student-related attributes such as attendance, assignment submission, assessment performance, course activity, login frequency, and interaction with learning resources. The system performs data preprocessing, feature selection, model training, and classification to identify students who may be at risk of dropping out. Machine learning algorithms such as Logistic Regression, Decision Tree, Random Forest, and Support Vector Machine can be evaluated using standard classification metrics. The proposed model can assist educators and administrators in identifying students who may require early academic support. It provides a data-driven decision-support approach for improving student monitoring and retention in online education.