基于多源数据的食品安全时空预警信息化体系设计研究

Design and investigation of food safety spatio-temporal early warning information system based on multi-source data

中文关键词:  食源性疾病  食品安全  机器学习  空间统计  时空预警

英文关键词:foodborne diseases  food safety  machine learning  spatial statistics  spatio-temporal early warning

中文摘要:

      大数据时代的来临, 机器学习技术不断被发掘, 时空分析技术日渐成熟, 两者的特点有机结合可以形成一种食品安全时空预警的新思路。本研究对上述技术进行了信息化层面的整合, 研究设计了基于跨部门多源数据的食品安全时空预警信息化体系框架, 利用机器学习对数据进行深度挖掘以及应用空间统计分析技术“建岛搭桥”, 在海量而复杂的监测数据中深度学习找到关键信息和隐藏关系, 对超出预期的风险异常区域进行预警, 为政府部门预防和控制食源性疾病提供科学依据和支持。

英文摘要:

      With the advent of the era of big data, machine learning technology has been constantly excavated, and space-time analysis technology is becoming more and more mature. The organic combination of the characteristics of the two technologies can form a new way of food safety space-time foodborne disease early warning. This research integrated the above technologies at the informationization level, and designed the framework of food safety information system based on large cross-sectoral data. In this study, machine learning was used to mine data in depth and spatial statistical analysis technology was used to build a bridge between islands. Key information and hidden relationship could be found in massive and complex monitoring data by in-depth learning, and risk abnormal areas could be early warned, which ccould provide government departments with foodborne diseases prevention and control, and give early warning of abnormal risk areas beyond expectations, so as to provide scientific basis and support for government departments to prevent and control foodborne diseases.