基于熵权法与改进的PCA聚类算法的电力客户价值分类与应用

Power consumer value classification and application based on entropy weight method and improved PCA clustering algorithm

对电力客户进行价值分析,有利于全面了解客户,为电力客户提供差异化服务。同时也可以提高客户满意度,实现供电企业、客户双赢的局面。运用数据挖掘方法对电力客户价值分类,构建电力客户价值评价指标体系,这些指标涵盖客户的用电行为、缴费行为、舆情和行业发展状况等,运用熵权法计算指标的权重,提出一种改进的PCA聚类算法对电力客户价值进行分类,为供电企业制定差异化服务策略提供辅助支撑。

The analysis of the power consumer value helps the power enterprises to understand their consumers comprehen- sively, provides the differentiated services for power consumers, and Can improve the consumer satisfaction to benefit both of the power supply enterprises and consumers. The data mining method is used to classify the power customer value, and con- struct the evaluation index system of the power consumer value. The indexes involve the consumer＇s power consumption behavior, paying behavior, public feelings and industrial development status. The weights of the indexes are calculated with the entropy weight method. A new improved PCA clustering algorithm is proposed to classify the power consumer value, which provides the ancillary support for the power enterprises to formulate the differentiated services strategy.