摘要
近年来,我国食品安全事件舆情监测的需求逐渐增加。针对该问题,笔者设计一个基于BERT-BiGRU多模集成的食品安全舆情分析系统。该系统通过爬虫技术抓取网络中针对食品安全事件的舆论文本数据,再调用一种基于BERT-BiGRU多模集成的深层情感语义识别方法进行情感分析,最后系统将分析后的结果在地图可视化、热力图等多个模块进行可视化展示。
In recent years, the demand for public opinion monitoring of food safety incidents is gradually growing in China. To address the issue,in this paper, we designed a food safety public opinion analysis system based on BERT-BiGRU multi-model ensemble Learning. The system crawls the text data of public opinion in the network through crawler technology, then invokes a deep sentiment semantic recognition method based on BERT-BiGRU multi-model ensemble learning for sentiment analysis, and finally the system visualizes the analyzed results in several modules such as map visualization and heat map.
作者
曹蕊
周毓奇
CAO Rui;ZHOU Yuqi(College of Computer and Information Engineering,Hubei University,Wuhan Hubei 430062,China)
出处
《信息与电脑》
2022年第7期94-97,共4页
Information & Computer
关键词
BERT预训练模型
双向门控循环单元
情感识别
舆情监测
集成学习
BERT pre-training model
bidirectional GRU
sentiment recognition
public opinion monitoring
ensemble learning
