An intelligent neuromarketing system for predicting consumers' future choice from electroencephalography signals

Fazla Rabbi Mashrur,Khandoker Mahmudur Rahman,Mohammad Tohidul Islam Miya,Ravi Vaidyanathan,Syed Ferhat Anwar,Farhana Sarker,Khondaker A Mamun
DOI: https://doi.org/10.1016/j.physbeh.2022.113847
2022-09-01
Abstract:Neuromarketing utilizes Brain-Computer Interface (BCI) technologies to provide insight into consumers responses on marketing stimuli. In order to achieve insight information, marketers spend about $400 billion annually on marketing, promotion, and advertisement using traditional marketing research tools. In addition, these tools like personal depth interviews, surveys, focus group discussions, etc. are expensive and frequently criticized for failing to extract actual consumer preferences. Neuromarketing, on the other hand, promises to overcome such constraints. In this work, an EEG-based neuromarketing framework is employed for predicting consumer future choice (affective attitude) while they view E-commerce products. After preprocessing, three types of features, namely, time, frequency, and time-frequency domain features are extracted. Then, wrapper-based Support Vector Machine-Recursive Feature Elimination (SVM-RFE) along with correlation bias reduction is used for feature selection. Lastly, we use SVM for categorizing positive affective attitude and negative affective attitude. Experiments show that the frontal cortex achieves the best accuracy of 98.67±2.98, 98±3.22, and 98.67±3.52 for 5-fold, 10-fold, and leave-one-subject-out (LOSO) respectively. In addition, among all the channels, Fz achieves best accuracy 90±7.81, 90.67±9.53, and 92.67±7.03 for 5-fold, 10-fold, and LOSO respectively. Subsequently, this work opens the door for implementing such a neuromarketing framework using consumer-grade devices in a real-life setting for marketers. As a result, it is evident that EEG-based neuromarketing technologies can assist brands and enterprises in forecasting future consumer preferences accurately. Hence, it will pave the way for the creation of an intelligent marketing assistive system for neuromarketing applications in future.
What problem does this paper attempt to address?