K-EmoPhone, A Mobile and Wearable Dataset with In-Situ Emotion, Stress, and Attention Labels

<em><strong>ABSTRACT: </strong></em>With the popularization of low-cost mobile and wearable sensors, many prior studies used such sensors to track and analyze people's mental well-being, productivity, and behavioral patterns. However, there is a lack of open datasets collected in real-world contexts with affective and cognitive state labels such as emotion, stress, and attention. This limits the advances in affective computing and human-computer interaction research. In this work, we present <em>K-EmoPhone</em>, an in-the-wild naturalistic dataset (n=80, 1-week) of smartphone use, wearable sensing, and self-reported affect states from college students. The dataset contains continuous probing of peripheral physiological signals and mobility data measured by off-the-shelf commercial devices in addition to context and interaction data by users' smartphones. Moreover, the dataset includes self-reports of in-situ affect states (n=5,753) such as emotion, stress level, attention level, and disturbance level, acquired by the experience sampling method. The resulting <em>K-EmoPhone</em> dataset helps to advance the research and development of affective computing, emotion intelligence technologies, and attention management based on mobile and wearable sensor data. Last update: Aug. 3, 2022 ----------------------------- * Version 1.0.0 (Aug. 3, 2022) Added <em>P##.zip</em> files, where each P## means the separate participant. Added <em>SubjData.zip</em> file, which includes individual characteristics information and labels.