摘要
目的构建适用于食品安全抽检数据的食品安全风险评估及预警系统。方法使用结构化查询语言,基于规则库引擎的高效匹配算法,通过对数据的预处理、构建基本事件模式和复合事件模式、规则库配置,对145个数据属性进行筛选优化。并使用53 047条食品安全抽检数据进行验证和测试。结果从大量属性中筛选出供系统使用的核心属性4类,形成基本规则7条,复合规则3条。通过验证,预警系统测试良好稳定,测试结果与人工标注结果一致。结论本研究创建了一种规则的生成和合成机制,建立了可实现实时预警和定时预警的食品安全风险评估及预警系统。该系统可实现食品安全风险评估的快速准确预警。
Objective To establish a food safety risk assessment and early warning system suitable for food safety sampling data. Methods By using structured query language, efficient matching algorithm based on rule base engine, 145 data attributes were filtered and optimized through preprocessing of data, construction of basic event mode and compound event mode, and rule base configuration. 53 047 pieces of food safety sampling data was used for verification and testing. Results 4 types of core attributes were selected, and 7 basic rules and 3 compound rules were formed. Through verification, the early warning system was good and stable, and the test results were consistent with the manual labeling results. Conclusion This study has created a rule generation and synthesis mechanism, and established a food safety risk assessment and early warning system that can realize real-time and early warning. The system can realize fast and accurate early warning of food safety risk assessment.
作者
王建新
王雅冬
闫利叶
王晔茹
WANG Jianxin;WANG Yadong;YAN Liye;WANG Yeru(School of Information,Beijing Forestry University,Beijing 100083,China;Engineering Research Center for Forestry-orented Intelligent Information Processing of National Administration,Beijing 100083,China;China National Center for Food Safety Risk Assessment,Beijing 100022,China)
出处
《中国食品卫生杂志》
CSCD
北大核心
2021年第1期1-7,共7页
Chinese Journal of Food Hygiene
基金
国家重点研发计划(2017YFC1602002,2018YFC1603302)
国家食品安全风险评估中心高层次人才队伍建设523项目。
关键词
食品安全
风险评估
风险预警
规则库引擎
智慧防控
大数据挖掘
Food safety
risk assessment
risk warning
rule base engine
smart prevention and control
big data mining