Research On Early Warning Model Of Wheat Quality And Safety Based On Data Mining

As a crop that is closely related to people’s daily life,wheat is a top priority directly related to consumers in its production,processing,transportation and other links of food safety.Wheat grains may be contaminated by various toxin-producing fungi during growth,harvesting,storage,and processing.Mycotoxins contamination is an important factor affecting the quality and safety of wheat crops.Taking a large number of sampling data of various mycotoxins in wheat by food supervision departments as the object,this paper studies the early warning model of wheat quality and safety by using data mining method.The main work includes:1.A method for early warning of wheat mycotoxins based on machine learning is presented.The method mainly includes a climate classification model of wheat mycotoxins samples and an AHP-LSTM early warning model based on wheat toxin monitoring data.Firstly,a climate classification model based on Smote-KNN was established according to the wheat toxin data;then,the quality risk early warning model of wheat mycotoxins was given by combining Analytic Hierarchy Process(AHP)and LSTM.The experimental results validate the method.2.An early warning method of mycotoxins contamination in wheat based on association rule mining is presented.Firstly,a method system for mining association rules of multiple mycotoxins contamination in wheat based on Apriori algorithm was constructed to analyze the characteristics of co-contamination of mycotoxins in wheat for early warning.Then,the Apriori algorithm was applied to the correlation analysis and early warning of multi-component mycotoxins in wheat,the monitoring data were classified into risk levels and the data were discretized,and the combined pollution of polytoxins in wheat was analyzed by mining strong association rules characterization.The experimental data visualization results verify the effectiveness of the method.3.The wheat quality and safety early warning system based on B / S mode has been designed and implemented.The model of front-end and back-end separation development has been used in wheat quality and safety early warning system,the back-end usesd the springboot framework of Java language and the mybatis framework,which are responsible for business logic processing and database operation.The front-end usesd HTML,CSS,Java Script and other technologies to complete the interaction.The main functional modules of the system include risk monitoring module,pollution level management module,risk early warning module,as well as the login module commonly used in the website system.