Journal Article

Customized ranking for products through online reviews: a method incorporating prospect theory with an improved VIKOR

Zhang, Chuan, Ling-wei, Fan, Ying-hui, Li, Yu-xin, Tian

2020

Overview

Online product reviews are significant in modern e-business because they can influence consumers’ purchase decisions. However, with the dramatic increase in the number of product categories and reviews, it is impossible for consumers to read all online reviews. In this paper, we design a novel method to help customers rank products using online reviews. Our method can be divided into three stages: generating a list of related alternative products based on specific filter conditions, collecting online reviews, and processing and measuring customer satisfaction. This study offers three significant improvements over previous approaches. First, we incorporate prospect theory that reflects the greater impact of negative reviews on customers’ purchase decisions to measure customer satisfaction concerning each attribute more accurately. Second, we combine the collective attribute weights calculated by entropy weight method (EWM) and the individual attribute weights given by a customer to improve VIKOR method, which can adjust the proportion of the two types of weights according to the customer’s knowledge of the product attributes. Third, for processing online reviews, we develop a new sentiment analysis algorithm that factors in the degree of consumer sentiment. This technique is different from the procedures used by existing studies for ranking products. To validate our method, we conduct a case study of automobile ranking and make some comparisons, which together demonstrate that the proposed method not only saves time and effort but also helps consumers select the products they really want.

• Publisher: Springer Nature B.V
• Subject: Algorithms / Consumers / Customer satisfaction / Data mining / Decisions / Electronic commerce / Product reviews / Ranking