基于数据驱动的多维度真烟异流预警研究
Research on Multidimensional Genuine Cigarette Diversion Warning Based on Data-Driven Approach
李峰 1张义忠 1黄种杰 1陈驰 1林观俊2
作者信息
- 1. 福建省烟草公司三明市公司,福建 三明 365000
- 2. 三明学院信息工程学院,福建 三明 365004
- 折叠
摘要
该文探讨了采用智慧专卖管理应用助手,通过融合卷烟投放策略数据、零售户信息、真烟外流数据、订单信息等数据,依据基本的概率理论和贝叶斯定理构建不同级别预警策略的方法.同时,基于真烟异流历史数据、零售户信息、预警信息等数据,该文分析了对零售户进行数字画像并通过星级评估策略量化零售户发生真烟异流风险的可行性.基于智慧专卖管理应用助手对多维度数据的综合、分析,可以为真烟异流的发生提供预测和依据,达到事前监管、精准查打的目的.
Abstract
This paper discusses the method of constructing warning strategies at different levels by using a smart monopoly management application assistant,integrating data such as cigarette distribution strategy,retailer information,genuine cigarette outflow data,and order information,based on basic probability theory and Bayes'theorem.Meanwhile,based on historical data of genuine cigarette diversion,retailer information,and warning information,this paper discusses the feasibility of creating a"digital portrait"of tobacco retailers and quantifying the risk of genuine cigarette diversion by retailers through a star-rating assessment strategy.According to the comprehensive and analytical application of multi-dimensional data by the smart monopoly management application assistant,it can provide prediction and basis for the occurrence of genuine cigarette diversion,achieving the purpose of ex-ante regulation and precise investigation.
关键词
数据驱动/烟草专卖管理应用助手/真烟异流/预警策略Key words
data-driven/tobacco monopoly management application assistant/genuine cigarette diversion/early warning strategy"引用本文复制引用
出版年
2024
NETL
NSTL
万方数据