基于张量分解的电网营销策略匹配算法研究
CSTR:
作者:
作者单位:

(1. 国网北京电力科学研究院,北京 100075;2. 北京恒华龙信数据科技有限公司,北京 100088)

作者简介:

张禄(1984),男,北京人,博士,研究方向为电动汽车数据分析与计算;李香龙(1980),男,河北保定人,硕士,研究方向为智能配用电技术。

通讯作者:

中图分类号:

TM714;F407.61

基金项目:

国家电网公司科技项目(520223170016)


Research on matching algorithm of power grid marketing strategy based on tensor decomposition
Author:
Affiliation:

(1. State Grid Beijing Electric Power Research Institute, Beijing 100075, China;2. Beijing Ibdpower Technology Co., Ltd., Beijing 100088, China)

Fund Project:

This work is supported by Science and Technology Project of State Grid Corporation(No. 520223170016)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    为了能够深入地探索电力客户特性,精确地掌握客户的服务需求,研究并制定了电力客户精细化营销策略,提升客户满意度。通过梳理精准营销研究理论,采用大数据挖掘技术分析客户的基本行为特性,构建一套符合满足客户特点和营销策略的客户属性指标体系,结合张量分解的多属性匹配算法,构建基于张量分解的多属性电力客户营销策略匹配模型,并以某电网公司1096户商业客户进行了实例分析,结果对电网公司开展精细化管理工作和采取差异化服务措施具有重要的参考价值。

    Abstract:

    In order to deeply explor the characteristics of power customers and accurately grasp the customers’service needs, refined marketing strategies for power customers are studied and formulated to improve customers’satisfaction. By combing the precision marketing theories, big data mining technology is adopted to analyze the basic behavior characteristics of customers, and a customer attribute index system that meets customer characteristics and marketing strategies is built. With the multi-attribute matching algorithm of tensor decomposition, the marketing strategy matching model of multi-attribute power customers based on tensor decomposition is constructed. An example analysis is carried out with 1 096 commercial customers of a power grid company. The results have important reference value for the power grid companies to carry out refined management work and differentiated services.

    参考文献
    相似文献
    引证文献
引用本文

张 禄,李香龙,朱 洁,陆斯悦,张宝群,马龙飞,陈少坤.基于张量分解的电网营销策略匹配算法研究[J].电力需求侧管理,2020,22(6):80-84

复制
分享
相关视频

文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2020-02-20
  • 最后修改日期:2020-09-25
  • 录用日期:
  • 在线发布日期: 2020-11-20
  • 出版日期:
文章二维码