Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

EEG-Based Consumer Behaviour Prediction: An Exploration from Classical Machine Learning to Graph Neural Networks

Mohammad Parsa Afshar, Aryan Azimi

TL;DR

The study tackles predicting consumer behavior from EEG signals by comparing classical machine learning, ensemble methods, and graph neural networks on the NeuMa neuromarketing dataset. It introduces three feature-pipeline strategies, a stacking ensemble, and brain-connectivity graphs to enable graph-based learning, while also systematically evaluating a wide range of classical models. Findings indicate that while classical and ensemble approaches often achieve higher overall accuracy, GNNs better capture minority-class signals by leveraging inter-electrode connectivity; combining these approaches may yield robust, balanced performance. The work demonstrates the promise of graph-based EEG modeling for neuromarketing and outlines practical directions for hybrid methods and multi-modal integration to enhance predictive power and interpretability.

Abstract

Prediction of consumer behavior is one of the important purposes in marketing, cognitive neuroscience, and human-computer interaction. The electroencephalography (EEG) data can help analyze the decision process by providing detailed information about the brain's neural activity. In this research, a comparative approach is utilized for predicting consumer behavior by EEG data. In the first step, the features of the EEG data from the NeuMa dataset were extracted and cleaned. For the Graph Neural Network (GNN) models, the brain connectivity features were created. Different machine learning models, such as classical models and Graph Neural Networks, are used and compared. The GNN models with different architectures are implemented to have a comprehensive comparison; furthermore, a wide range of classical models, such as ensemble models, are applied, which can be very helpful to show the difference and performance of each model on the dataset. Although the results did not show a significant difference overall, the GNN models generally performed better in some basic criteria where classical models were not satisfactory. This study not only shows that combining EEG signal analysis and machine learning models can provide an approach to deeper understanding of consumer behavior, but also provides a comprehensive comparison between the machine learning models that have been widely used in previous studies in the EEG-based neuromarketing such as Support Vector Machine (SVM), and the models which are not used or rarely used in the field, like Graph Neural Networks.

EEG-Based Consumer Behaviour Prediction: An Exploration from Classical Machine Learning to Graph Neural Networks

TL;DR

The study tackles predicting consumer behavior from EEG signals by comparing classical machine learning, ensemble methods, and graph neural networks on the NeuMa neuromarketing dataset. It introduces three feature-pipeline strategies, a stacking ensemble, and brain-connectivity graphs to enable graph-based learning, while also systematically evaluating a wide range of classical models. Findings indicate that while classical and ensemble approaches often achieve higher overall accuracy, GNNs better capture minority-class signals by leveraging inter-electrode connectivity; combining these approaches may yield robust, balanced performance. The work demonstrates the promise of graph-based EEG modeling for neuromarketing and outlines practical directions for hybrid methods and multi-modal integration to enhance predictive power and interpretability.

Abstract

Prediction of consumer behavior is one of the important purposes in marketing, cognitive neuroscience, and human-computer interaction. The electroencephalography (EEG) data can help analyze the decision process by providing detailed information about the brain's neural activity. In this research, a comparative approach is utilized for predicting consumer behavior by EEG data. In the first step, the features of the EEG data from the NeuMa dataset were extracted and cleaned. For the Graph Neural Network (GNN) models, the brain connectivity features were created. Different machine learning models, such as classical models and Graph Neural Networks, are used and compared. The GNN models with different architectures are implemented to have a comprehensive comparison; furthermore, a wide range of classical models, such as ensemble models, are applied, which can be very helpful to show the difference and performance of each model on the dataset. Although the results did not show a significant difference overall, the GNN models generally performed better in some basic criteria where classical models were not satisfactory. This study not only shows that combining EEG signal analysis and machine learning models can provide an approach to deeper understanding of consumer behavior, but also provides a comprehensive comparison between the machine learning models that have been widely used in previous studies in the EEG-based neuromarketing such as Support Vector Machine (SVM), and the models which are not used or rarely used in the field, like Graph Neural Networks.

Paper Structure

This paper contains 40 sections, 1 equation, 6 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Related works
  3. Datasets, stimuli, objectives, and task types
  4. Extracted features
  5. Models: Classical, Ensemble, and Neural Networks
  6. Research gaps and contributions of this research
  7. Methodology
  8. Dataset
  9. Classical Models
  10. Feature extraction
  11. Dimensionality Reduction and Data Preparation Pipelines
  12. Classical Models Descriptions
  13. Stacking Ensemble Models
  14. Graph Neural Network Models
  15. Feature Extraction
  16. ...and 25 more sections

Figures (6)

  • Figure 1: Overview of the proposed methodology for EEG-based consumer behavior prediction
  • Figure 2: Comparison of classical machine learning models with the PCA-based pipeline
  • Figure 3: Comparison of classical machine learning models with the UMAP-based pipeline
  • Figure 4: Comparison of classical machine learning models with the top features-based pipeline
  • Figure 5: Comparison of ensemble models with PCA-based, UMAP-based, and top features-based pipelines
  • ...and 1 more figures