Consumer Emotional State Evaluation Using EEG Based Emotion Recognition Using Deep Learning Approach

The standard methodologies for marketing (e.g., newspaper ads and tv commercials) are not effective in selling products as they do not excite the customers to buy any specific item. These methods of advertising try to ascertain their consumers’ attitude towards any product, which might not represent the actual behavior. So, the customer behavior is misunderstood by the advertisers and start-ups because the mindsets do not represent the buying behaviors of the consumers. Previous studies reflect that there is lack of experimental work done on classification and the prediction of their consumer emotional states. In this research, a strategy has been adopted to discover the customer emotional states by simply thinking about attributes and the power spectral density using EEG-based signals. The results revealed that, though the deep neural network (DNN) higher recall, greater precision, and accuracy compared with support vector machine (SVM) and k-nearest neighbor (k-NN), but random forest(RF) reaches values that were like deep learning on precisely the similar dataset.