International Research Journal of Education and Technology

In the evolving landscape of education, online assessments have gained significant traction, offering unparalleled flexibility and accessibility. However, these assessments pose challenges, particularly in accurate grading and identifying students needing additional support. This project proposes a web-based automatic test grading system that leverages decision tree algorithms to address these issues. By automating grading and integrating predictive analytics, the system provides educators with a robust tool to identify at-risk students and deliver tailored interventions effectively. The system operates by collecting and analysing student data, extracting meaningful features, and utilizing decision tree models to predict performance outcomes. Students identified as at-risk are assigned retests featuring simpler questions, designed to reinforce foundational concepts and provide them with an opportunity to improve. The decision tree algorithm is well-suited for this application due to its simplicity, interpretability, and ability to handle diverse data types. Moreover, its transparent decision-making process ensures educators can understand and trust the predictions, facilitating better-targeted support for students. By integrating this system into a web-based platform, educators can streamline test administration and grading while gaining actionable insights into student performance. The predictive and adaptive nature of the solution fosters an inclusive environment by addressing individual student needs, reducing educator workload, and enhancing overall academic outcomes. This innovative approach not only modernizes the assessment process but also contributes to a more equitable and effective educational experience.

Keywords: Online assessments, Automatic test grading, Decision tree algorithm, Student performance prediction ,  At-risk students, Machine learning in education Adaptive learning, Web-based grading system.