Stress and Affect detection using WESAD dataset Using WEarable Stress and Affect Detection (WESAD) dataset, training machine learning models such as Gaussian Mixture Model classifier, Random forests to classify Stress vs. Non-stress and also Stress vs. Neutral vs. Amusement.
Binary classification problem: Stress vs. Non-stress Logistic regression can be used to approximate non linear decision boundary.
3 class classification problem: Stress vs. Neutral vs. Amusement Clustering using GMM and classifying using GMM classifier, Training Random forests for classification
Different modalities: Chest vs. Wrist Evaluating the performance of the classifier based on different modalities
This dataset and idea is from the paper below: Philip Schmidt, Attila Reiss, Robert Duerichen, Claus Marberger, Kristof Van Laerhoven, "Introducing WESAD, a multimodal dataset for Wearable Stress and Affect Detection", ICMI 2018, Boulder, USA, 2018