数据与计算发展前沿 ›› 2024, Vol. 6 ›› Issue (5): 169-177.

CSTR: 32002.14.jfdc.CN10-1649/TP.2024.05.016

doi: 10.11871/jfdc.issn.2096-742X.2024.05.016

• • 上一篇    

基于多维度数据驱动的商品需求量预测研究

陈宇镔1(),洪烨1,崔文娟2,黄敏毅3,张锦玉2,*()   

  1. 1.广东烟草广州市有限公司，营销管理中心，广东 广州 510610
    2.中国科学院计算机网络信息中心，大数据部，北京 100083
    3.中国烟草总公司广东省公司，卷烟销售管理处，广东 广州 510610
  • 收稿日期:2023-12-08 出版日期:2024-10-20 发布日期:2024-10-21
  • 通讯作者: * 张锦玉（E-mail: jyzhang@cnic.cn）
  • 作者简介:陈宇镔，广东烟草广州市有限公司营销管理中心，副主任科员，长期从事卷烟产销计划制定、卷烟商品营销和卷烟市场研究工作，主要研究方向为卷烟市场供需状态分析、卷烟计划调控、卷烟商品市场策略制定等。
    负责初稿撰写、修改、审定与方向把握。
    CHEN Yubin, deputy director of Guangdong Tobacco Guangzhou Co., Ltd. Marketing Management Center. He has long been engaged in making production and sales plan for cigarette and cigarette market research. His research interests include cigarette market supply and demand status, cigarette supply adjustment and control, and cigarette marketing strategy.
    In this paper, he is responsible for the paper drafting, paper revision, and approval, as well as direction guidance.
    E-mail: 13084590@qq.com|张锦玉，中国科学院计算机网络信息中心，助理工程师，主要研究方向为机器学习、深度学习、神经网络、大数据分析。
    负责论文中数据分析与结果可视化。
    ZHANG Jinyu is a Junior Engineer at the Computer Network Information Center, Chinese Academy of Sciences. Her research interests include machine learning, deep learning, neural networks, and big data analysis.
    In this paper, she is responsible for data analysis and result visualization.
    E-mail: jyzhang@cnic.cn
  • 基金资助:
    中国烟草总公司广东省公司科技项目“基于多维度数据驱动的零售客户精准投放模型研究”(粤烟科项202111)

A Study on Multidimensional Data-Driven Commodity Demand Forecasting

CHEN Yubin1(),HONG Ye1,CUI Wenjuan2,HUANG Minyi3,ZHANG Jinyu2,*()   

  1. 1. Guangdong Tobacco Guangzhou Co., Ltd., Marketing Management Center, Guangzhou, Guangdong 510610, China
    2. Computer Network Information Center, Chinese Academy of Sciences, Big Data Department, Beijing 100083, China
    3. Guangdong Provincial Company of China Tobacco Corporation, Cigarette Sales Management Office, Guangzhou, Guangdong 510610, China
  • Received:2023-12-08 Online:2024-10-20 Published:2024-10-21

RichHTML

12

PDF

178

摘要：

【目的】面向烟草行业，为改善零售户商品资源分配策略，优化货源供应政策，解决依据历史销售水平和经验分配商品时存在的缺少数据理论体系支撑、货源分配不合理、货源供应政策落后和业务操作效率低等问题。【方法】提出了多维数据驱动的神经网络模型，数据包括商户内部订单数据和外部商圈等，通过对数据进行异常样本删除、特征构造和划分训练验证集后，在训练集上依次通过Attention层、Dense层和LightGBM模型。【结果】最终在测试集上实现了96.57%的预测准确度。【结论】该技术基于卷烟系统多维数据，能够建立具备高度适应性与灵活性的需求量预测体系，打破现行按档位确定需求量的操作模式，实现高精度的预测准确率，为企业卷烟智能投放提供技术支持。

关键词: 神经网络, 需求预测, 特征构造, Attention层, LightGBM模型

Abstract:

[Objective] For the tobacco industry, it is crucial to improve the commodity resource allocation strategy of retailers, optimize the supply policy of commodities, and solve the problems of unreasonable allocation of goods, behindhand commodities supply policy and low efficiency of business operations, which exist in the current commodities allocation based on the historical sales level and experiences but lack of systematic data theory support [Methods] A multi-dimensional data-driven neural network model is proposed. the data used by the model includes merchant internal order data and external shopping area, etc. By abnormal sample removal, feature construction, and partition of training and validation data sets, the attention layer, the dense layer, and the LightGBM model layer are passed successively on the training set. [Results] A prediction accuracy of 96.57% is finally achieved on the test set. [Conclusion] Based on the multi-dimensional data of the cigarette system, this technology can establish a highly adaptable and flexible demand forecasting system, break the current operation mode of determining the demand according to the stalls, realize high prediction accuracy, and provide technical support for the intelligent placement of cigarettes for enterprises.

Key words: neural networks, demand forecasting, feature construction, Attention layer, LightGBM model