Research And Application Of Data Mining Technology In Food Agricultural Products Safety Early Warning

Food safety early warning system is of great significance to the realization of food safety strategy.It is an indispensable means to prevent food safety problems and strengthen food safety supervision.The food safety early warning system can provide the government and market-related personnel with scientific auxiliary decision-making information,reduce losses caused by agricultural product safety issues,in a sense,it can also play a role in maintaining social stability.Based on data mining technology,mining and analyzing agricultural product testing data in agricultural product wholesale markets,building agricultural product safety early warning models and developing a set of agricultural product safety early warning systems have important application value.The main research contents of the project and the phased results obtained are as follows.(1)Research on the association relationship of unqualified agricultural products based on Apriori algorithm.Firstly,in order to improve the data quality and provide guarantee for the performance of subsequent data mining algorithms,it is necessary to process the data set of agricultural products,including cleaning,specification,and transformation of agricultural product data,Secondly,mining the association relationship between different attributes of unqualified agricultural products through the association relationship mining algorithm Apriori,and exploring the relationship between unqualified agricultural products.The two types of relationships mined provide basic data for the early warning model of agricultural product safety.(2)Research on the construction of agricultural product safety early warning model.Firstly,taking agricultural products as research objects,mining the association relationship between the attributes of unqualified agricultural products according to Apriori,the mining results are generated into the first component of agricultural product safety warning information.Secondly,taking the merchants in the agricultural product market as the research object,according to Apriori,mining the association relationship between different unqualified agricultural products to obtain merchant information related to agricultural products and conduct research on merchant credit evaluation models in the agricultural product market.Merchant credit evaluation model includes two parts: credit evaluation index system and index weight.Determining merchant credit evaluation indicators according to the actual needs of the agricultural product market,and carrying research on the weight value of credit evaluation index based on the Entropy method.And then constructing a sorted list of safety warnings according to the merchant credit evaluation index based on the Entropy method--TOPSIS algorithm.Finally,three warning levels are set according to the entropy value-the characteristics of the TOPSIS algorithm and actual needs of agricultural product market management,which composes the second part of agricultural product safety early warning information.Through two parts of agricultural product safety early warning information,the construction of agricultural product safety early warning model is realized.(3)Design and implement the early warning system of agricultural products safety.Using the agricultural product safety early warning model proposed in this paper,and according to the actual needs of the market,carrying out the system requirement analysis,system outline design and system detailed design according to the software engineering method,and then the early warning system of agricultural product safety is realized.