Analysis of Consumer Social Media Representation Based on Large Language Models:A Case Study of Taobao's Return Service
In the digital age,business managers need to actively listen to consumer feedback on products and services through big data technology in order to continuously improve their operational performance.Social media serves as an important channel for consumers to freely share and discuss their perceptions of products and services after shopping.Analyzing,summarizing,and refining the vast amount of messy text information continuously generated by consumers to obtain the core expressions of consumer groups is a challenging task.This study proposes an analytical framework for social media analysis based on large language models and social representation theory,and uses Sina Weibo tweets to analyze Taobao return services as an example to illustrate the specific application of this framework.Case analysis reveals that return difficulties,product quality,logistics distribution,packaging issues,etc.,are core issues reflected by consumers.Among them,return difficulties and product quality issues often accompany more negative emotions from consumers.Compared with traditional topic analysis techniques,the new analytical framework can help enterprises refine multi-dimensional and multi-level social representations from consumer social media data,obtain more accurate,in-depth,and comprehensive insights into consumer perceptions on their products and services.
Social Media AnalysisLarge Language ModelsSocial RepresentationSentiment AnalysisReturn Services
NETL
NSTL
万方数据
