EduCore: An AI-integrated classroom management system with automated grading and academic integrity detection for university-level education

The rapid advancement of artificial intelligence (AI) in education presents significant opportunities to enhance classroom management, automate assessment processes, and promote academic integrity. This study presented EduCore, a full-stack web and mobile classroom management system developed using Next.js and integrated with the Anthropic Claude API (claude-3-5-sonnet-20241022). EduCore introduced a three-tier role-based access architecture encompassing school administrators, teachers, and students, with an approval-based enrollment pipeline designed to ensure institutional control and security. The system incorporated a configurable grading engine adaptable to diverse university grading policies, supported multiple submission types including written essays, multiple-choice quizzes, coding assignments, and file uploads, and integrated AI-powered student performance feedback subject to teacher approval before release. A key feature of EduCore was its AI-generated content detection mechanism, which analyzed submitted academic work to flag potential non-original outputs. The system was evaluated by a general university population composed of students and teachers through a structured survey instrument using a five-point Likert scale based on the ISO 25010 Software Quality Model. Results indicated high levels of usability, functionality, and acceptability across both user groups. This study contributed an empirically validated, AI-integrated classroom management framework relevant to the evolving landscape of higher education technology.