Intelligent analysis of EEG signals to assess consumer decisions: A Study on Neuromarketing
This work addresses the problem of improving marketing strategies for businesses by using neuromarketing techniques, but it is incremental as it applies existing machine learning methods to new EEG data.
The study tackled the problem of assessing consumer decisions by analyzing EEG signals to understand positive and negative reactions to advertisements and products, achieving accuracies up to 0.63 with methods like SVM and NB in subject-dependent analysis and finding deep learning models competitive in product and ads-based analysis.
Neuromarketing is an emerging field that combines neuroscience and marketing to understand the factors that influence consumer decisions better. The study proposes a method to understand consumers' positive and negative reactions to advertisements (ads) and products by analysing electroencephalogram (EEG) signals. These signals are recorded using a low-cost single electrode headset from volunteers belonging to the ages 18-22. A detailed subject dependent (SD) and subject independent (SI) analysis was performed employing machine learning methods like Naive Bayes (NB), Support Vector Machine (SVM), k-nearest neighbour and Decision Tree and the proposed deep learning (DL) model. SVM and NB yielded an accuracy (Acc.) of 0.63 for the SD analysis. In SI analysis, SVM performed better for the advertisement, product and gender-based analysis. Furthermore, the performance of the DL model was on par with that of SVM, especially, in product and ads-based analysis.