基于关联规则挖掘的食品安全信息预警模型  

The Information Pre-warning Model of Food Safety Based on Association Rules Mining

作　　者：顾小林[1,2] 张大为[3] 张可[4] 浦徐进 曹文彬 

机构地区：[1]河海大学,南京210098 [2]江南大学,无锡214122 [3]辽宁师范大学,大连116029 [4]重庆大学,重庆400030

出　　处：《软科学》2011年第11期136-141,共6页Soft Science

基　　金：国家自然科学基金项目(70901034);江南大学人文社会科学预研项目(2010wyy005);江南大学自主科研计划学科交叉创新团队基金项目(JUSRP31107);江南大学江苏省食品安全研究基地项目(10SPJD006);中央高校基本科研业务费专项资金资助项目(JUSRP21101)

摘　　要：针对食品生产加工的安全问题,建立了基于关联规则挖掘的食品安全信息预警模型,此模型从数据挖掘视角以食品生产加工的检测数据为处理对象,分析数据异常情况,采用改进的关联规则挖掘APTPPA算法,生成关联路径树、寻找频繁项集、最大频繁项和最大关联规则,进而抽取报警关联规则,挖掘导致食品安全问题的因素并进行诊断和预警,最后通过实验验证了预警挖掘模型的正确性和算法的有效性。This study builds the food security information pre-warning modal based on association rules mining against the security problems of food production and processing. The model takes the data of food production and processing as handling objects from the perspective of data mining. It analyzes the abnormal data and adopts the improved association rules mining algorithm of APTPPA, generates associated path tree, and finds frequent itemsets, maximal frequent itemsets and maximal association rules. Then it extracts alarm association rules to mine the factors that lead to food safety problems and to diagnose and pre-warning. Finally, experiments verify the correctness of the model and the effectiveness of the algorithm.

关 键 词：关联规则 APRIORI算法 APTPPA算法 食品安全预警模型 

分 类 号：F208[经济管理—国民经济]