摘要
针对现有技术分析客户用电行为速度慢、效率低下的问题,提出了新型的大数据算法。该方案采用混合大数据算法实现营销大数据客户用电行为的多种分析,在Apriori算法模型的技术上融入改进型K-means均值聚类算法,通过采用极限学习机(Extreme Learning Machine,ELM)原理,实现营销大数据客户用电行为的快速分类,提高了分类能力。通过数据关联挖掘,准确地发现用户用电行为与影响因素之间潜在关系,提高了用电分析的能力。试验表明,该研究方案准确率高达93%。
Aiming at the problems of slow and inefficient in analyzing customers’electricity consumption behavior by the existing technologies,a new type of big data algorithm is proposed.The solution uses hybrid big data algorithms to realize the electricity consumption behavior of marketing big data customers,integrates the improved K-means clustering algorithm into the technology of the Apriori algorithm model,and adopts the principle of Extreme Learning Machine(ELM),realizes the rapid classification of marketing big data customers’electricity consumption behavior,and the classification ability is improved.Through data association mining,the potential relationship between users’electricity consumption behavior and influencing factors is accurately discovered,which improves the ability of electricity consumption analysis.Tests show that the accuracy of this research scheme is as high as 93%.
作者
王俊
戴璐平
冯秀庆
潘晔
WANG Jun;DAI Lu-ping;FENG Xiu-qing;PAN Ye(State Grid Shanghai Municipal Electric Power Company,Shanghai 200120,China)
出处
《信息技术》
2021年第4期125-129,135,共6页
Information Technology
