Oran, SametGursel, Amira Tandirovic2025-01-062025-01-062021978-605011437-910.23919/ELECO54474.2021.96778572-s2.0-85125277871https://doi.org/10.23919/ELECO54474.2021.9677857https://hdl.handle.net/20.500.14669/128913th International Conference on Electrical and Electronics Engineering, ELECO 2021 -- 25 November 2021 through 27 November 2021 -- Virtual, Bursa -- 176537In this work, EEG signals that are showing the frequency of the power bands were examined by wavelet power spectrum through neuromarketing outline in order to predict purchaser appetites while they look E-commerce goods. When extraction of these power bands, fixed overlap segment and 3 different sample lengths were counted in sliding window technique. k-NN was implemented for evaluating classification accuracy. The best result, 70.24% k-NN accuracy was obtained for 2-seconds sample length. © 2021 Chamber of Turkish Electrical Engineers.eninfo:eu-repo/semantics/closedAccessEEG Classificationk-NNNeuro-marketingRandom ForestSliding WindowWavelet SpectrumConference Object228224