Statistical Analysis of Electroencephalographic Signals in the Stimulation of Energy Data Visualizations

Increasing luxury living standards coupled with technological developments have made energy efficiency in homes much more important. By protecting the environment and preventing the depletion of energy resources, making energy use conscious has an important role in preserving a livable world for future generations. The brain-computer interface (BCI) has been widely used to improve the quality of life of individuals. There have been numerous research projects on predicting the behavior of energy consumers. However, detecting emotional responses and incorporating personal perceptions of individuals is still a challenge to understanding energy users' behavior. This paper investigates the best selection method for energy data visualization using BCI systems for improving energy users' experience. An experimental study has been conducted to acquire electroencephalography (EEG) signals of energy users against the stimuli of energy data visualizations to detect emotions and understand the users' perceptions. A self-assessment manikin (SAM) is used to rate the arousal and valence scales required for emotion classification. Sample entropy (SampEn) and approximate entropy (ApEn) are utilized to analyze EEG data. One-way ANOVA and Tukey's honestly significant difference test is applied to the entropy values of EEG signals showing some promising results from the conducted statistical analysis.