Liking Status Estimation Using Eeg Signals and Mode Decomposition Method

Abstract

Emotion estimation is a very important and popular research area from human-computer interaction point of view. In this study, a liking estimation method is proposed by using electroencephalogram (EEG) signals for neuromarketing applications. The liking status is estimated by using signal processing and machine learning methods applied to EEG recordings taken while the subjects watched the advertisement videos of two different car brands. After viewing videos, the participants were asked to give ratings from 1 to 5 while they are presented the picture of 7 car sections (front view, console, side view, rear view, stop lamp, logo and front grille) taken from the advertisement video. EEG segments corresponding to these regions were analyzed by Multivariate Empirical Mode Decomposition (MEMD) method. Various features were calculated from the extracted intrinsic mode functions (IMF) and liking status classification was performed. The successful results show that the proposed MEMD-based method may be used in neuromarketing studies. © 2022 IEEE.