• PASUMARTHI AMRUTHA
• DARAPU UMA

Online Judge (OJ) tools are extensively utilized in programming educational institutions due to their ability to evaluate student submissions swiftly and impartially. This evaluation method typically results in a singular outcome derived from a rubric that often indicates whether the submission has met the assignment's requirements. However, it would be beneficial for both educators and learners to have a greater influence over the overall assessment of the project, as this automated evaluation may not fully capture the nuances of academic performance. This initiative aims to tackle this concern by optimizing the use of OJ data and providing prompt feedback to both teachers and students. By employing learning-based methodologies to simulate student behavior, including Multi-Instance Learning and fundamental machine learning techniques, the accuracy of assessments can be improved. Additionally, Explainable AI is being explored to furnish individuals with comprehensible feedback. The hypothesis was evaluated through a case study involving 2,500 submissions from approximately 90 students enrolled in a computer science program with a focus on programming. The model's ability to accurately predict a student's outcome, such as the likelihood of passing or failing an assignment based solely on behavioral patterns identified in OJ submissions, lends credence to the hypothesis. Moreover, this approach can uncover additional pertinent information, such as student profiles and at-risk groups, thereby facilitating feedback for both educators and learners.

Online Judge, Explainable AI, Assignment, Feedback.

"EXPLAINABLE AI: PROVIDING STUDENT FEEDBACK VIA ONLINE JUDGE", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 8, page no.f443-f453, August-2024, Available :http://www.jetir.org/papers/JETIR2408659.pdf

"EXPLAINABLE AI: PROVIDING STUDENT FEEDBACK VIA ONLINE JUDGE", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 8, page no. ppf443-f453, August-2024, Available at : http://www.jetir.org/papers/JETIR2408659.pdf