Food affection determination through biosignal based affective computing

Abstract

When people seek food, their emotions shift, and not all foods are appealing in all moods. It is incredibly tough to learn people's eating habits and provide advice depending on their emotions at the time. The goal of this study is to create a system that will categorize emotional information from meals based on a person's present emotion. This study proposes a technique for evaluating a person's emotion using an electroencephalogram (EEG) output. To that purpose, we organized trials for EEG data collecting as well as questionnaires. On the raw data, a prominent feature extraction approach known as Discrete Wavelet Transform (DWT) was utilized, and Gradient Boosting Classifier and AdaBoost Classifier were used for affectivity classification. We obtained a notable Accuracy Score and AUC Score from these classifiers in this study.