基于数据挖掘技术的电信集团客户占有率应用研究

Application Research of Telecom Customers Share Based on Data Mining

随着电信行业集团业务的高速发展,一般的基于客户保有的统计算法已不能满足集团业务的需要。为了更好地把握集团业务方向以及对集团业务的管理与开拓,提出了在数据仓库与数据集市下,基于数据挖掘技术的占有率统计方法,首先利用K-means算法产生分类得出真实率,再在真实率的基础上求得占有率,最后通过真实率和占有率能够有效地分析当前集团业务的总体概况。这种方法为决策者提供了真实、直观的数据依据,极大地弥补了客户保有算法的不足。

With the fast growth of telecom industry, the general statistical algorithm based on customer retention can no longer meet the needs of the telecom business. In order to better grasp the direction of business, and manage and promote the business, this paper proposes the statistical method based on data mining in the data warehouse and data mart. Firstly, it use K-means algorithm to generate classification and get the real rate, and then get the occupation ratio on the basis of real rate, finally, analyzes the current overview of telecom business by means of real rate and occupation ratio effectively. This approach can provide true and intuitive date to decision makers. In addition, it completely compensates the lack of customer retention algorithm.