Construction of consumer satisfaction evaluation index system for green products based on online comments

Abstract

The promotion of green product consumption and its transformation towards low-carbon alternatives is essential for implementing new developmental paradigms and achieving carbon neutrality objectives. This study employs text mining techniques to analyze user online comments from e-commerce platforms, focusing on consumer satisfaction regarding green products. Utilizing the KeyBert model, relevant keywords were extracted from user feedback, followed by the training of keyword vectors using Word2Vec. K-means clustering was then employed to develop a comprehensive system of consumer satisfaction evaluation index system for energy-saving air conditioning products on the JD platform. The findings reveal that consumers prioritize functionality, service quality, aesthetic appeal, pricing, logistics, and installation in their evaluations. It is recommended that manufacturers enhance installation procedures, refine aesthetic designs, and emphasize functional advantages to elevate consumer satisfaction. However, this study is limited by its focus on a singular product category and necessitates further research incorporating a broader dataset to validate these findings. Future investigations should consider a wider range of green products and leverage diverse data sources to enrich the analysis.