Neural Network-based Classification of EEG Signals for Predicting Consumer Decision-Making

The instrument known as EEG (Electroencephalogram) is utilized to examine brain activity during various tasks. These signals are helpful in identifying brain processes and movements, making them valuable in detecting neurological disorders. The pulses recorded have a low magnitude, ranging from ۰.۵ to ۱۰۰ V and are measured from peak to peak. This is significantly smaller than ECG signals, making EEG signals susceptible to noise. It is crucial to preprocess these signals accurately for precise assessment and detection of brain illnesses. The EEG is a combination of signals from different small locations on the cortical plate beneath the electrodes, located at the crown of the head. These electrical impulses can be used to evaluate a wide range of brain diseases such as hypertension, Alzheimer's, and brain damage. By studying the EEG signal, we can compare and differentiate brainwaves associated with different emotions and illnesses. Several research studies have been conducted to develop methods for preprocessing, feature extraction, and evaluation of EEG data. In recent years, there has been an emerging trend in using EEG for human-computer communication, which is a challenging and innovative field. Our report presents a predictive model for analyzing customers' preferences and dislikes through EEG signals. These impulses were recorded while the participants were browsing the internet for various consumer goods.

Keywords:

EEG (Electroencephalogram) , analyzing customers , Bio-acoustic features , features’ reproducibility , speech signal processing , speech task