摘要
针对流调中重点人员的隐私保护问题,文章利用联邦学习技术,提出了一种基于联邦学习和隐私保护的流调分析方法,构建了相应的流调分析系统,目的是在数据保留在本地不直接共享的条件下,各机构进行数据共享和联合建模,得出重点人员的车次信息、行动轨迹,精准呈现近期的活动区域,将这些轨迹和区域在地图中标明,帮助有关部门及时制定对策并开展防控工作,提高时效性,避免因划定区域太过广泛而加大防疫工作量。
Aiming at the key personnel privacy protection in the epidemiological survey and analysis, this paper, by using the learning technology, is proposed based on a federal study and epidemiological analysis of privacy protection method, builds the corresponding epidemiological survey and analysis system, which is aimed at data retention under the condition of local not directly sharing, data sharing and joint modeling agencies, concluding the key personnel train number information, action, and accurately presenting the recent activity areas and marking these tracks and regions on the map as to help relevant departments formulate countermeasures and carry out prevention and control work in a timely manner, improve timeliness, and avoid increasing epidemic prevention work due to too extensive demarcations.
作者
廖晓婷
胡明
LIAO Xiaoting;HU Ming(Hubei Posts and Telecommunications Planning and Design Co.,Ltd.;Wuhan Branch of China United Network Communications Co.,Ltd.,Wuhan,Hubei,430023)
出处
《长江信息通信》
2022年第6期169-171,共3页
Changjiang Information & Communications
关键词
联邦学习
隐私保护
流调分析
新冠疫情
运动轨迹
Federated learning
Privacy protection
Epidemiological investigation and analysis
COVID-19 pandemic
Movement trajectory
