Design And Implementation Of Quality Risk Monitoring Sysyem For Imported Products Based On Public Opinion

In recent years,with the improvement of people's living standard,people's demand for overseas shopping are increasing quickly.And the scale of cross-border electronic commerce are also developing rapidly.When people enjoy high-quality goods,great pressure has been brought to the work of inspection and quarantine of imported products.Now,the prediction of the quality risk of imported products mainly depends on the historical statistical data,which is not enough to deal with emergencies.With the popularity of the Internet,the importance of network public opinion gradually emerged.In the field of inspection and quarantine,public opinion mainly includes the notice issued by the government department,the announcement issued by the organization and the user's comments on the electronic business platform.Network public opinion by virtue of its good timeliness,can effectively compensate for the current inspection and quarantine methods and help for the relevant departments to provide decision.The quality risk monitoring system of imported products discussed in this paper comprehensively utilizes structured historical statistical data and unstructured network public opinion data to achieve risk prediction of imported products and other related business needs.It provides powerful decision-making for the relevant departments.The work done in this paper is as follows:1.Introduce the background and significance of risk prediction of imported products,and expound the research status of related technologies at home and abroad.2.Introduce the principle of analytic hierarchy process,use the analytic hierarchy process to combine public opinion rules with traditional rules.Taking the risk prediction of imported food products as an example,introduce the detail of process of modeling.3.The two key modules search engine and the rule engine are discussed in detail:? search engine module.The system uses a distributed full-text search framework,which can search and count text data in real-time under massive data,and provide data support for risk prediction and front-end display.? Rule engine module.The system uses different rule chains to match different products according to different types of products.The match result is assigned to the risk prediction model to calculate the risk index and risk level.Finally,use the knowledge base to reason the risk.