基于BERT模型的微博文本细粒度情感分析

打开文本图片集

中图分类号：TP391.4 文献标识码：A 文章编号：2096-4706（2026）04-0112-05

Fine-grained Sentiment Analysis of Weibo Text Based on BERT Model

ZHANGYimin，LIYe （SchoolofInformationScienceand Technology，Shanghai Sanda University，hanghai2O2O9，China）

Abstract： With therapid development of social media，Weibo，as an important platform for user information exchange andemotionexpressionaccumulatesmassiveandrichtextdata.Basedontheinvestigationofrelated technologies forWeibo sentiment analysis，this paper proposesa fine-grained sentiment analysis methodbasedonthe BERTmodel.Combining Weibo datacrawlingand preprocessingtechnologies，thispaperconstructsanefientanalysis frameworkThismethodcompletesdata colectionthrough the WeiboOpenAPI，utilizes the BERTpre-training model torealizetext vectorization，andcompletes the preciseclassatioofsxemotiosiuingager，apiss，tralusess，neabedotor architecture.Meanwhile，this paper introduces sentiment dictionaries anddataaugmentation technologies to improve model performance，and displays analysis results by means of visualization tools.The study shows thatthis method achieves high classification accuracy on the SMP2020 dataset and provides a new idea for fine-grained sentiment analysis of Weibo text.

Keywords：Weibo sentiment analysis; fine-grained sentiment; BERT model; social media

0 引言

随着社交媒体的快速发展，微博作为中国最具影响力的社交平台之一，已成为公众情感表达和信息传播的重要渠道。（剩余6559字）