Proceedings of the 4th International Conference on Economic Management and Model Engineering, ICEMME 2022, November 18-20, 2022, Nanjing, China

Research Article

Study on Consumer Acceptance Evaluation Model of Ceramic Tea Set Based on Facial Emotion Recognition Technology

Download166 downloads
  • @INPROCEEDINGS{10.4108/eai.18-11-2022.2326848,
        author={Xianghui  Li and Takaya  Yuizono and Ruixuan  Li and Bing  Wang},
        title={Study on Consumer Acceptance Evaluation Model of Ceramic Tea Set Based on Facial Emotion Recognition Technology},
        proceedings={Proceedings of the 4th International Conference on Economic Management and Model Engineering, ICEMME 2022, November 18-20, 2022, Nanjing, China},
        publisher={EAI},
        proceedings_a={ICEMME},
        year={2023},
        month={2},
        keywords={ceramic tea sets facial emotion recognition classifiers machine learning features of tea sets},
        doi={10.4108/eai.18-11-2022.2326848}
    }
    
  • Xianghui Li
    Takaya Yuizono
    Ruixuan Li
    Bing Wang
    Year: 2023
    Study on Consumer Acceptance Evaluation Model of Ceramic Tea Set Based on Facial Emotion Recognition Technology
    ICEMME
    EAI
    DOI: 10.4108/eai.18-11-2022.2326848
Xianghui Li1,*, Takaya Yuizono2, Ruixuan Li1, Bing Wang3
  • 1: School of Art and Design, Dalian Polytechnic University, Dalian, China
  • 2: Graduate School of Advanced Science and Technology, Japan Advanced Institute of Science and Technology, Nomi, Ishikawa, Japan
  • 3: Landscape Design Studio, China Northeast Architecture Design and Research Institute Co., Ltd, Dalian, China
*Contact email: dalian2006@163.com

Abstract

The purpose of this study is to explore a consumer acceptance evaluation model of ceramic tea sets based on facial emotion recognition and machine learning. Through the evaluation experiment of ceramic tea sets, we obtained the facial expression pictures of participants when observing tea sets and the questionnaire data of Self-assessment Manikin (SAM). Then the Microsoft facial emotion recognition API (Microsoft azure) was used to calculate the facial expressions of participants and form a data set of facial emotion characteristic variables. Finally, the models of consumer acceptance recognition were built by using random forest (RF) and neural network (NN) classifier. The results show that: 1) the accuracy of emotion value recognition using RF classifier was 82.26% and that of arousal is 74.7%. These two results are better than the model built by the NN classifier. This shows that the RF classifier model can meet the market acceptance evaluation of ceramic tea sets in practice, and contribute to the design and development of ceramic tea sets. 2) The participants in this experiment have a higher acceptance of delicate, shiny, and beautiful tea sets, while they gave a lower score for tea sets with primitive and crude style, rough texture, and heavy shape, which also confirms the general aesthetic taste of young people.