| 摘 要: 为了对社会突发事件引发的舆情及时开展分析、引导及治理,在大数据环境下,采用SSM(Spring+SpringMVC+MyBatis)架构设计并实现了一个集微博舆情信息采集、去重、分析、处理及可视化的综合平台,同时对微博舆情分析的数据处理进行了介绍,并对文本情感分析、计算文本相似度给出了具体算法。系统通过数据采集层、数据处理层和数据展示层3个子层实现对微博舆情信息的趋势性预警及有效监管。目前,该平台完成了相关微博舆情热点话题的追踪及趋势预警,在实际应用中,观察到在单机日采集量约100万条时,对舆情数据分析的有效率可以达到90%以上。 |
| 关键词: 网络舆情;大数据;微博 |
|
中图分类号: TP319
文献标识码: A
|
| 基金项目: 广东省自然科学基金项目(2021A1515011803);广东省哲学社会科学规划学科共建项目(GD18XXW07). |
|
| Design and Implementation of Weibo Public Opinion Analysis |
|
HUO Ying1, QIU Zhimin2, LI Xiaofan1, LI Yanting1
|
(1.School of In f ormation Engineering, Shaoguan University, Shaoguan 512005, China;
2.School of Intelligent Engineering, Shaoguan University, Shaoguan 512005, China)
huoying@sgu.edu.cn; 250437325@qq.com; 14929099@qq.com; kidi@qq.com
|
| Abstract: In order to timely analyze, guide, and manage public opinion triggered by social emergencies,this paper proposes to adopt SSM (Spring + SpringMVC + MyBatis) architecture to design and implement a comprehensive platform for collecting, de-repeating, analyzing, processing and visualization of Weibo public opinion information under the big data environment. At the same time, the data processing of public opinion analysis is introduced, and the specific algorithms of text sentiment analysis and text similarity calculation are also introduced. The system realizes trend warning and effective supervision of Weibo public opinion information through three sub-layers: data acquisition layer, data processing layer and data display layer. At present, the tracking and trend warning of relevant hot topics of public opinion on Weibo have been completed on the platform. In practical application, it is observed that the efficiency of public opinion data analysis can reach more than 90% when the daily collection volume of a single machine is about 1 million. |
| Keywords: Internet public opinion; big data; Weibo |
