EvalSeer: An Intelligent Gamified System for Programming Assignments Assessment

摘要

Continuous evaluation of computer programs and providing informative assessments are crucial for computer programming students. However, swift and formative feedback can be challenging to achieve as it is usually a stressful and tedious task for professors merely through manual grading. There is an urgent need for a Learning management system (LMS) that offers instant and detailed feedback in a competitive environment for a better education experience. In this study, we introduce the EvalSeer learning management system. EvalSeer is an LMS equipped with an intelligent auto-grading engine to keep learners motivated and help them move forward. The code evaluation process covers various criteria that strengthen coding abilities and provides learners with the directions they need to improve. These criteria include coding style, code features, dynamic test cases, and successful compilation. EvalSeer uses Long short-term memory (LSTM) networks for code analysis to detect syntax errors and predict potential fixes. Also, the system shall explain suggested fixes backed up with related references. EvalSeer is an easy-to-use cloud-based system with a learner-first approach that can be applied both on-campus and in elearning systems. This work is timely with the dramatic education change, with a notable rise of e-learning due to the COVID-19 pandemic.