EEG Dataset Collected During Energy Data Visualization Stimuli Presentation (EDAVIS)

Published: 24 October 2023| Version 3 | DOI: 10.17632/w9nk4mvgbb.3


User behavior has a significant impact on household energy consumption. Though researchers use a variety of methods to investigate user behavior, the solutions for evaluating user behavior are limited. This article presents an open-access electroencephalography (EEG) dataset that contains EEG data from individuals stimulated by energy data visualizations. The dataset includes 28 healthy participants' 6-channel EEG recordings. A 32-channel EMOTIV EEG device is utilized to acquire EEG signals, and an international 10-20 electrode system is employed to place electrodes. The stimuli are created and presented using PsychoPy software. Through the use of a self-assessment manikin (SAM), participants rate the valence and arousal of each stimulus to determine their affective state for that stimulus. Additionally, three questions are asked for each stimulus. The dataset includes original data visualizations and ratings. To facilitate analysis, the raw EEG data is segmented into data visualizations and neutral images using event markers. Using the EMOTIVPro application, EEG recordings are directly saved, and the PsychoPy application is used to store subjective responses. Artificial EEG data is also produced using Generative Adversarial Networks (GANs) for the enhancement of data size and classification performances. This dataset suggests a novel application of EEG research and offers a helpful starting point for researchers in the fields of computer science, energy efficiency, artificial intelligence, brain-computer interfaces, and human-computer interaction.



De Montfort University


Artificial Intelligence, Energy Efficiency, Data Visualization, Electroencephalography, Human-Computer Interaction, Brain-Computer Interface, Generative Adversarial Network, Data Augmentation