基于Apriori算法的供电公司营销数据挖掘系统设计
DOI:
作者:
作者单位:

1.国网天津市电力公司;2.国网天津城东供电分公司;3.天津市普迅电力信息技术有限公司

作者简介:

通讯作者:

中图分类号:

TP391

基金项目:


Design of marketing data mining system for power supply company based on Apriori algorithmZhang Jian1, Liu Chang2, Yang Yi Wei2 Xinzhe2, Zhang Hao2, Wang Xu3
Author:
Affiliation:

Fund Project:

  • 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:

    随着供电公司累积数据量的不断增加,海量数据的处理与管理难度逐渐提升,为了保证高质量数据,提供更为周到的营销服务,提高执行效率,设计基于Apriori算法的供电公司营销数据挖掘系统。在硬件设计上,通过营销数据挖掘系统物理架构部署,搭建系统硬件环境,实现数据库信息的同步。在软件设计上,设计电力营销数据仓库,收集整理原始业务数据,抽取电力营销数据,采用Apriori算法,通过映射剪枝处理频繁项集,计算置信度,挖掘关联规则,以此建立多维数据挖掘模型,选择数据挖掘空间,计算各维度的贡献分量,实现系统的数据挖掘功能。经实验论证分析,本文系统在电力负荷预测应用中的预测结果与实际值相差较小,在最小支持度和事务数据量条件下,数据挖掘执行时间分别在2s和10s以下,具有较高的执行效率,说明本文系统是可行的。

    Abstract:

    With the continuous increase of accumulated data of power supply companies, the processing and management of massive data is becoming more difficult. In order to ensure high-quality data, provide more thoughtful marketing services, and improve the execution efficiency, a power supply company marketing data mining system based on Apriori algorithm is designed. In terms of hardware design, through the deployment of the physical architecture of the marketing data mining system, the system hardware environment is built to achieve the synchronization of database information. In terms of software design, design the power marketing data warehouse, collect and sort out the original business data, extract the power marketing data, adopt the Apriori algorithm, process frequent item sets by mapping pruning, calculate confidence, and mine association rules, so as to establish a multidimensional data mining model, select data mining space, calculate the contribution components of each dimension, and realize the data mining function of the system. Through experimental demonstration and analysis, the difference between the predicted results and the actual values of the system in the application of power load forecasting is small. Under the conditions of minimum support and transaction data volume, the execution time of data mining is less than 2s and 10s respectively, which has high execution efficiency, indicating that the system in this paper is feasible.

    参考文献
    相似文献
    引证文献
引用本文
分享
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:2022-11-17
  • 最后修改日期:2023-03-23
  • 录用日期:2022-11-29
  • 在线发布日期:
  • 出版日期: