A Product Recommendation System Based on User Complaint Analysis Using Product Reviews
Terutaka Yoshikawa, Yuanyuan Wang, Yukiko Kawai · 2019
In recent years, e-commerce attracts attention to the spread of online shopping sites, many users who refer to product review comments to choose the products and higher-rated products can be recommended from users' purchase histories. However, traditional recommendation systems cannot recommend alternative products when existing products' reviews contain user complaints (e.g., price, battery, design, function, etc), it is still difficult to satisfy users' requirements. Therefore, in this work, we propose a novel product recommendation system that analyzes two kinds of information: Complaint information and satisfaction information from review comments on e-commerce. It is possible to recommend alternative products to satisfy your requirements that can solve the complaint information on the product when you browsing. In this paper, we describe the complaint information extraction based on product review analysis by extracting both negative information and positive information from product reviews, and we also explain alternative product recommendation method that can solve complaints, and we verify the effectiveness of complaint information extraction and alternative product recommendation.