AI-Enabled LMS for Advanced and Flexible Academic Management: A Comparative Study with Experimental Evaluation

Abstract

This study addresses the challenge of making digital learning platforms, known as Learning Management Systems (LMS), more adaptive and efficient. We investigated whether artificial intelligence (AI) could create a more flexible system that lightens teacher’s workload while personalizing the student experience. To answer this, we developed an AI-powered LMS and conducted a comparative evaluation against traditional methods. Our system provides three key contributions: (1) using machine learning to auto-generate adaptable course syllabi, (2) applying natural language processing to automatically grade written essays, and (3) leveraging predictive analytics to tailor individual learning paths for each student. We tested this platform with 500 university students across various subjects. The results demonstrated clear advantages. Instructors using the AI tools created courses nearly 70% faster. In grading, the AI was not only faster. It provides scores in seconds but also slightly more consistent and achieving 93% accuracy compared to 85% for human graders, with a 91% agreement rate between the two. Furthermore, student satisfaction scores saw a significant jump from 72% to 88%. Our discussion also tackles critical implications, such as ensuring algorithmic fairness and defining the appropriate role for human oversight in AI-assisted grading. These findings confirm that a thoughtfully designed AI can significantly enhance education by handling repetitive tasks, thereby freeing teachers to focus on more impactful interactions. Based on these results, our future work directly extends from our findings: we are now exploring AI-driven assessment of broader, real-world skills through multiple data formats, and investigating blockchain to create tamper-proof credential verification, building on the need for trustworthy and comprehensive automated systems.