"十四五"时期，烟草行业面临社会环境发生巨大变革的挑战。从行业现状来说，烟草市场面临消费需求日益多元化、市场竞争日趋激烈、销售结构提升矛盾突出等问题。随着数字技术的不断发展，数据驱动逐渐成为烟草行业的新推手。零售户数据作为最基本的数据来源，可以帮助企业有针对性地优化市场布局。本文以河南中烟CRM客户管理系统中的零售户信息为数据基础，以CRISP-DM(Cross Industry Standard Process forData Mining)为研究框架，结合逻辑回归、ARIMA时间序列、BP神经网络等机器学习和深度学习模型，对黄金叶(天叶)规格卷烟在2021年第四季度的销售数据进行建模验证，助力精准把控未来零售户价值走向。
Application of Clustering Algorithm in Retailer Value Analysis
During the 14th Five Year Plan period,the tobacco industry is facing the challenge of significant changes in the social environment.From the current situation of the industry,the tobacco market is facing increasingly diversified consumer demand,fierce market competition,and prominent contradictions in improving sales structure.With the continuous development of digital technology,data-driven has gradually become a new driving force in the tobacco industry.Retail data,as the most basic source of data,can help businesses optimize their market layout in a targeted manner.This study is based on the retail information in the Henan Zhongyan CRM customer management system,using CRISP-DM as the research framework,and combining machine learning and deep learning models such as logistic regression,ARIMA time series,BP neural network,etc.,to model and verify the sales data of Huangjinye Tianye specification cigarettes in the fourth quarter of 2021.On the basis of verifying the accuracy of the model,accurately control the future value direction of retail households.
cigarette marketvalue forecastingCRISP-DM data mininglogical regression model
万方数据
维普
