The Multi-EMP dataset aims to collect natural interactions in the remote online environment during the conversation process between people and measure the level of empathy between the two. Based on the definition of 'Empathy' in psychology, empathy in this study focuses on the sharing of emotions between each other. In this study, two people engage in a conversation in four sessions, and all interactions are recorded through video, audio, and wearable devices to collect physiological signals. In each session, participants engage in a conversation for 8 minutes, responding to emotions and the intensity of emotions every 2 minutes. Based on these response points, the recorded videos are segmented. Thus, 528 videos are provided with visual, audio, and text data, and each participant's EDA, BVP, TEMP, and MET are provided as minute-by-minute data. In addition, two levels of empathy are provided: the empathy level calculated based on the collected data and the empathy level responded to by the listener.
Access Request
To access the dataset and download the files, please send an email to 218354@jnu.ac.kr, showing a motive to use the dataset for non-commercial academic research purpose. The download link will be provided via email.
Eunchae Lim, Hwaryung Lee, Ji-eun Shin, Hyung-Jeong Yang, Soo-Hyung Kim, Seungwon Kim, Aera Kim, "Multi-modal adaptive empathy assessment in online dyadic interaction using bi-directional multi-layer perceptron-mixer and dynamic weights fusion", Engineering Applications of Artificial Intelligence, Vol. 167, 113864, 2026
This work was supported in part by the National Research Founda tion of Korea (NRF) grant funded by the Korea Government [Ministry of Science and ICT (MSIT)] under Grant RS-2023-00219107(50%), in part by the Institute of Information and Communications Tech nology Planning and Evaluation (IITP), South Korea through the Artificial Intelligence Convergence Innovation Human Resources De velopment grant funded by the Korea Government (MSIT) under Grant IITP-2023-RS-2023-00256629(30%), and in part by the IITP Innovative Human Resource Development for Local Intellectualization program grant funded by the Korea government (MSIT) under Grant IITP-2026-RS-2024-00437718(20%).