Personalized Stress Detection using Multimodal Dataset from Wearable Sensor Fitri Indra Indikawati
Universitas Ahmad Dahlan
Abstract
Stress detection is an interesting topic because of its huge impact on human health, both mentally and physically. Physiological changes on the human body can be observed and used to recognize stress. Various approaches utilize a different kind of data to detect physiological changes, i.e. visual data, audio data, interview result, textual data, or sensor data. Wearable sensors for collecting physiological data are becoming more prominent in recent years due to their functionality and non-intrusive nature. By utilizing data from wearable sensors, we have developed a personalized stress detection system. Our system performs classification on stress level based on multimodal data from Empatica E4 wearable sensor. By evaluating the performance of the system, we demonstrate that our system can perform personalized stress detection using a real multimodal dataset from wearable sensor.
