
Study of Neuromarketing: Visual Influence with Decision Making on Impulse Buying
Author(s) -
Rifat Januar,
Hilman Fauzi,
Maya Ariyanti,
Faradisya Heris
Publication year - 2021
Publication title -
kinetik
Language(s) - English
Resource type - Journals
eISSN - 2503-2267
pISSN - 2503-2259
DOI - 10.22219/kinetik.v6i4.1334
Subject(s) - neuromarketing , computer science , electroencephalography , stimulus (psychology) , pattern recognition (psychology) , artificial intelligence , psychology , neuroscience , cognitive psychology
Marketing trends have been increasing in the last few decades. Products need good branding and the right marketing strategy. Various marketing methods have been widely done, and one of them is with the study of neuroscience, especially neuromarketing. Neuromarketing is used to seek the influence of marketing stimuli on consumers and objective data through advances in neurology by utilizing human senses such as restraint, smell, taste, and touch. Measurements of neuromarketing responses to the brain can use electroencephalography signals (EEG). Measurement is done with the visual stimulus of consumers when making decisions. To analyze consumer interests, the majority still using qualitative methods, but it is still considered less effective due to many uncertain factors. In this study, neuromarketing responses were measured to the human brain using (EEG) signal analysis. Data collection was conducted on 11 respondents with a stimulus in the form of different product colors and was affected by changes in light intensity. For pre-processing used bandpass filters to get beta signals in the absence of noise. Then the data will be processed using Fast Fourier Transform (FFT) and energy extraction as characteristic extraction and classification of Support Vector Machines (SVM) in the signal pattern recognition process. The results of testing the best feature combination parameters showed an accuracy value of 72% with a combination of magnitude and phase features. By using the range of phase feature values obtained an accuracy of 67% for signal pattern recognition respondents.