一种海关品牌型号库构建工具设计方法

A novel design method of building tool for customs brand model library

目前海关常用的十位商品海关编码是一种比较粗放的编码方式,在报关单中商品海关编码很难和具体商品实现一一对应的关系,亟待构建一套标准的商品智能分类工具,为海关管理提供基准的商品分析依据。文章基于海关报关单申报商品名称规格型号,利用机器学习等先进技术手段,实现将海关商品的最小分析单元从HS编码细化至具体品牌型号,形成商品基础库,实现商品智能分类,为贸易监管、税收征管、稽查缉私等业务提供强大支撑。

At present,10 bit commodity HS code is a rough coding method commonly used by Customs.Unfortunately it is difficult to achieve a one-to-one correspondence between the commodity HS code in the customs declaration and the specific commodity,which lead to urgently build a set of standard intelligent commodity classification system to provide the basis of commodity analysis for customs management.According to the customs declaration of commodity name,specification,model,using advanced technology such as machine learning,this paper refines the minimum analysis unit of customs commodity from HS code to specific brand and model,forming a basic commodity library,realizing intelligent classification of commodities,which provide strong support for supervising trade,tax collection and management,anti-smuggling and other businesses.