面向应急管理的事故和灾害网络舆情监管系统

NETWORK PUBLIC OPINION MONITORING SYSTEM OF ACCIDENTS AND DISASTERS FOR EMERGENCY MANAGEMENT

系统基于Hadoop框架,利用MapReduce编程模型,采用K-means聚类算法,结合实际舆情处置工作流程,将舆情监测、处理、分析、上传下达等功能融为一体。面向应急管理服务,针对主要灾害及政务类网络信息开展全天候监测,第一时间获取事件与灾害"原生态"舆情数据,并实现热点话题的快速识别与跟踪处理。通过省-市-区三级用户协同管理,在事件发生后快速发现相关信息与可能存在的网络舆论危机,分析网络舆论倾向,提供分级处置接口,为及时、有效、全面掌握与处理突发事件网络舆情提供技术支撑。

Based on Hadoop framework,we build the system that integrates public opinion monitoring,processing,analysis,uploading and other functions by adopting MapReduce programming model and K-means clustering algorithm,and by combining with the actual public opinion disposal workflow.For emergency management services,the system could monitor network information throughout the day for major disasters and government network information.Therefore,we could obtain the original public opinion data of events and disasters at the first time and identify and track the hot topics rapidly.Through the collaborative management of users at the provincial,municipal and district levels,the relevant information and possible network public opinion crisis are quickly found after the event.The system can analyze the trends of network public opinion and provide hierarchical disposal interface,which provides technical support for timely,effective and comprehensive mastery and handling of public opinion in emergencies.