基于C4.5算法分类器的电力客户信用评级模型研究

发布时间：2018-07-01 10:34

  本文选题：C.算法 + 决策树　； 参考：《通化师范学院学报》2015年12期

【摘要】：电力客户巨额欠费是电力公司面临的一个难题,该文采用C4.5算法分类器对电力系统客户进行信用评级.从电力系统大规模的缴费数据中提取训练样本,利用C4.5算法进行学习得到分类规则,然后将这些规则应用于用电客户的信誉评级,从而更好地管理客户缴费行为,为电力系统的管理运营提供数据支持.实验结果表明,对于不同的指标体系和不同的分类样本,都可以获得较好的分类效果.
[Abstract]:The huge arrears of electricity customers is a difficult problem faced by power companies. In this paper, C4.5 algorithm classifier is used for credit rating of power system customers. Training samples are extracted from large-scale payment data of power system, and C4.5 algorithm is used to learn the classification rules. Then, these rules are applied to the credit rating of electricity customers, so as to better manage the customer's contribution behavior. Provide data support for power system management and operation. The experimental results show that better classification results can be obtained for different index systems and different classification samples.
【作者单位】： 安徽工商职业学院;
【基金】：国家自然科学基金面上项目“基于灵敏性分析和隐因素发现的复杂系统脆弱性演化机制研究”(61175051)
【分类号】：F426.61;F274
，

本文编号：2087498