A Product Recommendation Method by Analyzing Sales Volume, Sales Period, and User Satisfaction
Haoyang Xia, Yuanyuan Wang · 2024
In recent years, the large amount of product information available on e-commerce websites has made it difficult for users to refer to it when shopping and selecting products. In particular, the comments posted by customers on e-commerce websites usually reflect the users' impressions and feelings about using the products objectively and realistically. Therefore, in order to provide guidance to customers who need to buy the products, we propose a product recommendation method based on the three critical dimensions of the product: sales volume, sales period, and user satisfaction. To do this, we use the sentiment analysis of product reviews to calculate the satisfaction score of the product. Then, this satisfaction score is combined with the sales volume and sales period of the product to calculate the comprehensive ranking score to provide users with a product ranking for recommending products with high satisfaction scores, large sales volumes, and long sales periods. Finally, we evaluate and discuss the proposed method using actual e-commerce data.