Abstract:
This report presents a novel framework for real-time stress detection using wearable devices, focusing on Electrodermal Activity (EDA). Current datasets lack accurate annotations, and existing methods struggle in ambulatory settings due to resource constraints. Our framework addresses these issues by continuously monitoring EDA and prompting users for quick annotations upon stress detection, improving data accuracy and enabling personalized interventions for stress management.
