Group interaction and evolution of customer reviews based on opinion dynamics towards product redesign

  • PDF / 1,382,048 Bytes
  • 21 Pages / 439.37 x 666.142 pts Page_size
  • 3 Downloads / 187 Views

DOWNLOAD

REPORT


Group interaction and evolution of customer reviews based on opinion dynamics towards product redesign Fan Zou1   · Yupeng Li1 · Jiahuan Huang1 Accepted: 6 November 2020 © Springer Science+Business Media, LLC, part of Springer Nature 2020

Abstract Perception of customer requirements and intention is crucial for product redesign where customer reviews play a significant role. Customers dynamically make decision and interact with others, which lead to the evolution of customer reviews. A customer reviews evolution model (CREM) is proposed to analyse the dynamic evolution process of group customer reviews by using a modified Deffuant-Weisbuch model based on opinion dynamics. In the proposed methodology, negativity bias and the helpfulness of reviews are incorporated according to the characteristics of customers and reality. Based on the related literature reviews and survey, negativity bias is introduced to present that positive customers are still sensitive to negative reviews out of confidence radius and will interact with them. In addition, the helpfulness of reviews is used to reflect the rate of information acquisition since the ability of expression varies from person to person. Moreover, as a case study, the customer reviews evolution of a smartphone is modelled to support the redesigned attributes evaluation. Finally, the feasibility and effectiveness of the proposed CREM is expounded through result analysis and discussion. Keywords  Customer reviews · Opinion dynamics · Negativity bias · DeffuantWeisbuch model · Product redesign

* Fan Zou [email protected] Yupeng Li [email protected] Jiahuan Huang [email protected] 1



Department of Industrial Engineering, School of Mines, China University of Mining and Technology, Xuzhou, Jiangsu, China

13

Vol.:(0123456789)



F. Zou et al.

1 Introduction With the development of customer-oriented market, identifying customer requirements becomes more and more significant. Especially for the product of famous enterprise, an inappropriate product development scheme may result in customer dissatisfaction and even customer loss. Due to the tendency of customization and the innovative pressure from markets, product redesign is regarded as a crucial process for modern industry, which can focus on the improvement of the specific part and prevent extra cost and time from the unnecessary changes [1–3]. Moreover, as one of the impulses of redesign, customer requirement is the critical guideline since it implies the market feedback [4–6]. Thus, any launch of a redesigned product needs to match requirements of customers as well as a new product. In product development stage, many techniques are used to identify and predict customer requirement and behavior including data mining, machine learning, and cloud computing, which are applied to seize customer purchase information and analyze customer behavior rules [7–10]. In contrast with these methods that acquire customer requirement indirectly, customer reviews that positively present individual attitude, preference and requireme