A study on kansei attraction of products’ online reviews by using text mining and kano model

Abstract

Customers' emotional needs, also called Kansei demands, have become one of the most focuses in new product development (NPD). With the rapid growth of the Internet of Things, customers are pleased to share their emotional experience and preference for products through an online platform. However, how to excavate customers' potential real needs in massive online reviews is the key to NPD. In order to better recognize and satisfy customers' emotional needs, this study proposes to explore the Kansei attraction of online products in combination with text mining and Kano model. Firstly, text mining technology extracts useful Kansei information from massive customer online reviews data. Then Kano model investigates the interaction between product Kansei and customer satisfaction, determines the Kansei attractive quality that greatly enhances customer satisfaction, and successfully predicts the future trend of products. These emotional qualities provide valuable references for enterprises, and designers can derive corresponding product design features based on them, which will improve the success rate of new product launches. A case study of extracting slow juicer's online reviews from Amazon.com is used to demonstrate the feasibility of the method and the results also can be extended to other NPD.