Abstract:

Student engagement and timely communication remain persistent challenges in higher education, particularly within hybrid and remote learning environments. Traditional Learning Management System (LMS) notifications are often static and confined to in-platform delivery, resulting in missed deadlines, reduced participation, and communication fatigue. This paper proposes a conceptual framework for an AI-driven adaptive multi-channel notification system designed to enhance engagement by personalizing how and when academic reminders are delivered. The methodology integrates learning analytics and behavioral modeling to analyze student interaction patterns, including response timing, platform usage, and engagement frequency. Predictive algorithms identify each student’s most effective communication channels, such as email, mobile messaging, or LMS alerts, along with optimal delivery times. Notifications are dynamically routed and continuously refined through a feedback loop that adapts to evolving behaviors. A simulated implementation is used to evaluate the framework. Anticipated results include measurable improvements in assignment submission rates, increased course participation, reduced response latency to academic prompts, and higher overall student responsiveness. The model is also expected to decrease notification overload by optimizing message timing and channel selection without increasing total communication volume. This research contributes a scalable, data-informed approach to Artificial Intelligence in Education (AIED), offering both practical and theoretical insights for improving digital engagement in blended and online learning environments.

Keywords:

Adaptive Systems, Artificial Intelligence in Education, Learning Analytics, Student Engagement, Hybrid Learning, Educational Technology.