Classification System Based on Electrodermal Activity Using Supervised Learning Algorithm with Galvanic Skin Response

This research focuses on the development of an emotion classification system based on electrodermal activity using a supervised learning algorithm with Galvanic Skin Response. The study addresses the growing concern over mental health disorders in Indonesia, which highlights the need for technology based tools to assist medical professionals in objectively identifying patient emotions. The classification method employed is Support Vector Machine, trained to recognize three primary emotional states: happy, sad, and angry. The data processing pipeline includes data cleaning, normalization, and statistical feature extraction to enhance model accuracy. The model achieved a training accuracy of 96.9 percent and a testing accuracy of 92 percent in classifying the targeted emotions.