摘要
为提高数据溯源算法的运行效率,基于特征提取方法提出电力客户服务大数据溯源模型。定义数据的基因组,查找数据库内的任意两个存在血缘关系的数据,以此建立数据染色体追溯模型;获取电力数据中的四类特征值,建立电力数据多次遗传的转移矩阵,基于特征提取构造电力大数据溯源路径;得出大数据溯源算法,构建电力客户服务大数据溯源模型。实验结果显示,特征提取算法在模型层数以及数据量相同时,溯源所需时间最短,算法运行速度最快。
In order to improve the efficiency of data traceability algorithm,a power customer service big data traceability model is proposed based on automatic feature extraction method.It defines the genome of the data,searchs any two data in the database with blood relationship,and establishes the data chromosome traceability model.Four kinds of characteristic values in power data were obtained,and the transfer matrix of power data multiple genetic was established,and the power big data traceability path was constructed based on automatic feature extraction.The big data traceability algorithm is obtained and the power customer service big data traceability model is constructed.The experimental results show that the automatic feature extraction algorithm has the same number of model layers and data volume,the shortest time required for traceability,and the fastest running speed.
作者
于亮
钟宏伟
李海涛
刘志欣
苏姗姗
YU Liang;ZHONG Hong-wei;LI Hai-tao;LIU Zhi-xin;SU Shan-shan(State Grid Beijing Electric Power Company Customer Service Center,Beijing 100078 China)
出处
《自动化技术与应用》
2024年第9期101-104,共4页
Techniques of Automation and Applications
关键词
特征提取
电力客户服务
数据库
大数据
数据溯源算法
feature extraction
power customer service
database
big data
data traceability algorithm
作者简介
于亮(1980-),女,硕士,高级工程师,研究方向:电力营销。
