Personalized Product Ranking Based on Linguistic Requirements and Online Reviews

Abstract

With the rapid development of online shopping, an increasing number of consumers purchase products and share reviews on e-commerce platforms. Various methods have been proposed to assist consumers in making purchasing decisions based on online product reviews. However, most existing methods focus on objective product aspects, general consumer preferences, or historical user preferences, neglecting the current preferences of specific users. This paper proposes a personalized product ranking method based on the linguistic requirements of current users and historical online reviews, under the framework of multi-attribute decision-making. This method extracts attributes and their weights from the linguistic requirements, taking the current preferences of specific users into account, and increasing flexibility. Furthermore, it segments online reviews based on attribute transitions to improve the accuracy of product attribute scores derived from these online reviews. A case study and comparative analysis are conducted to verify the effectiveness of this proposed method, demonstrating its ability to incorporate the current preferences of specific users and increase the accuracy of product attribute scoring based on online reviews.