AI-Based Application for Task Management and Scheduling Student Activity

Abstract

University students face significant challenges in managing academic demands, which often lead to procrastination, stress, and diminished mental well-being. To address this, we developed a proactive AI-based assistant designed to support student productivity and health. The application leverages Large Language Models (LLMs) and an agentic AI framework based on the ReAct pattern to offer personalized task prioritization, dynamic scheduling, and cognitive load reduction. It uniquely integrates academic, emotional, and biological factors, such as circadian rhythms, to provide holistic, context-aware support. A usability study involving 30 participants showed favorable outcomes, achieving a System Usability Scale (SUS) score of 73.67, a Task Success Rate (TSR) of 83% for the AI scheduling task, average Single Ease Question (SEQ) score of 5.61 on a 7-point scale (where higher is better) indicating that the system is good enough to use. Qualitative feedback highlighted user satisfaction with the system's stability and AI-driven scheduling capabilities. This research presents a novel, adaptive platform that shifts from reactive, siloed educational tools to an anticipatory support system. The findings validate the potential of agentic AI to enhance academic performance, offering a scalable model for future student support in higher education.

Ähnliche Arbeiten