Imported Appliance Risk Level Identification Based on Support Vector Machine Algorithm

Abstract

Since 2009, China has promulgated several laws and regulations to regulate the import of solid waste, but there has been a lack of supporting identification criteria. To provide detailed and feasible risk level identification criteria for imported appliances to guide the Customs identification of e-waste. This paper establishes a three-tier identification criterion which has 42 indicators covering: appearance, value of use, electrical safety risk, mechanical safety risk, toxic and hazardous substances risk. Using these indicators as input, an intelligent identification method constructed by support vector machine (SVM) algorithm could identify the risk level of imported appliances as low risk, medium risk, and high risk. To verify the effectiveness and practicality of this method, this paper uses the identification cases provided by Wuxi Customs. The results show that the identification method has high self-learning capability and accuracy.