Supervised Learning for EEG-Driven Consumer Preference Prediction in Functional Food Consumption

Obesity and diabetes are leading contributors to cardiovascular disease, driven largely by ultra-processed diets. In this work, we propose a deep learning framework to analyze EEG patterns in functional food consumption, using Principal Component Analysis (PCA) to extract features and a neural network architecture to model consumer decision-making. Our supervised learning approach predicts taste preferences for a unique consumer group without prior ratings, demonstrating that decision making can be effectively modeled using neural signals. By optimizing the hyperparameters, we achieve a 1.23% improvement in the efficiency of the model, providing a scalable solution to understand the diet choices. This work bridges neuroscience and machine learning, offering information on consumer behavior with potential applications in personalized nutrition and health-aware food recommendation systems.