<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>In this paper, we first extracted the time-domain features, frequency-domain features and spatial-domain features of EEG signals, combined with the three-stage feature selection algorithm applicable to the binary classification problem and the multi-classification problem, and constructed the SEE model for emotion recognition based on EEG signals. Then, based on the three-level design model of emotion, emotion decoding and labeling are carried out on the instinctive layer, behavioral layer and reflective layer of product design, and the constructed model is combined to improve the product design emotionally. Finally, after analyzing the results of product emotion annotation, we explore the performance of the EEG-based emotion recognition model and the improvement effect of product design emotionalization. The results showed that the average accuracy of the EEG signal emotion recognition model for various emotion recognition was about 0.99, and the intensity of emotion intensity in Dahe was 0.32 and 0.25, respectively, accounting for 0.57 of the total sample, and the performance evaluation indicators of the eight emotions were greater than 0.85. Ninety percent of product experiencers had pre- and post-improvement differences between [0.12, 0.22] for happiness and [-0.20, -0.04] for dissatisfaction.</p>
</abstract>

In this paper, we first extracted the time-domain features, frequency-domain features and spatial-domain features of EEG signals, combined with the three-stage feature selection algorithm applicable to the binary classification problem and the multi-classification problem, and constructed the SEE model for emotion recognition based on EEG signals. Then, based on the three-level design model of emotion, emotion decoding and labeling are carried out on the instinctive layer, behavioral layer and reflective layer of product design, and the constructed model is combined to improve the product design emotionally. Finally, after analyzing the results of product emotion annotation, we explore the performance of the EEG-based emotion recognition model and the improvement effect of product design emotionalization. The results showed that the average accuracy of the EEG signal emotion recognition model for various emotion recognition was about 0.99, and the intensity of emotion intensity in Dahe was 0.32 and 0.25, respectively, accounting for 0.57 of the total sample, and the performance evaluation indicators of the eight emotions were greater than 0.85. Ninety percent of product experiencers had pre- and post-improvement differences between [0.12, 0.22] for happiness and [-0.20, -0.04] for dissatisfaction.

Research on Emotional Improvement of Product Design Based on Emotion Recognition Technology

XING ZHI COLLEGE OF XI’AN UNIVERSITY OF FINANCE AND ECONOMICS

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

In this paper, we first extracted the time-domain features, frequency-domain features and spatial-domain features of EEG signals, combined with the three-stage...

Research on Emotional Improvement of Product Design Based on Emotion Recognition Technology

Data publikacji: 05 gru 2023

Zakres stron: -

Otrzymano: 14 mar 2023

Przyjęty: 15 cze 2023

DOI: https://doi.org/10.2478/amns.2023.2.01357

Słowa kluczowe
Electroencephalographic signals, Feature selection algorithm, SEE model, Three levels of emotion, Product design emotionalization

© 2023 Lujuan Xin, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Research on Emotional Improvement of Product Design Based on Emotion Recognition Technology

Data publikacji: 05 gru 2023

Zakres stron: -

Otrzymano: 14 mar 2023

Przyjęty: 15 cze 2023

DOI: https://doi.org/10.2478/amns.2023.2.01357

Słowa kluczoweElectroencephalographic signals, Feature selection algorithm, SEE model, Three levels of emotion, Product design emotionalization

© 2023 Lujuan Xin, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Słowa kluczowe
Electroencephalographic signals, Feature selection algorithm, SEE model, Three levels of emotion, Product design emotionalization