【作者】 白爱民；

【导师】 齐二石；

【作者基本信息】 天津大学 ， 管理科学与工程， 2006， 博士

【摘要】 鉴于CRM技术的迅速发展以及它在实际中的广泛应用,CRM的理论和研究正在成为管理科学中的一个热点的研究课题。本文在研究了大量国内外相关文献的基础上,以CRM在某省级电信企业的理论与应用为研究背景,对CRM的发展及其相关理论、CRM模型的建立、内外部机制、关键算法以及CRM应用系统等题进行了较深入地探讨。1.针对CRM的客户集群理论,通过对集群→企业集群→客户集群三重概念的剖析,构建了适合于CRM的理论模型框架;对集群概念及其涵义进行了扩展和引申,建立了CRM体系结构。在对外部机制研究的基础上,对其内部核心机制的规律及运作模式进行了深入地探讨。2.采用拓扑理论方法,研究了CRM客户集群理论,构建了客户集群的数学模型,提出了企业集群、客户集群、信息集群的拓扑空间概念;并对该空间的内涵进行了深入地分析,阐述了客户集群的基、集族、子空间与映射等实体概念。3.提出了CRM客户集群核心机制的四种客户联系模式,根据客户联系的映射原理,将该理论模式与电信企业的营销实体应用进行了对应的映射分析。4.对CRM数据挖掘过程的关键环节——数据预处理存在问题和算法进行了研究。提出了一种基于密度的孤立点因子算法和一种基于粗集理论的属性类别差异数据归约算法。对CRM数据预处理与数据挖掘的孤立点异常检测方法,对采取粗集理论的预处理过程的数据约简与处理算法进行了分析和编程。5.结合省级电信企业在CRM系统应用项目的实施,对上述理论进行了实证研究,以验证理论的正确性,并给出其实际应用效果。该项研究和应用取得了显著的社会和经济效益,并具有应用和推广价值。更多还原

【Abstract】 The rapid development of CRM technology and its application in practice, the theory of CRM and its research is becoming a hot problem in the field of science of management. Based on a large number of research of references,and on the background of provincial application research of CRM, the development of CRM, the modeling of CRM , internal and external mechanism, and CRM application system was deeply researched.1. Aim to customer cluster of CRM, and through the analysis of the three layer concept: cluster→enterprise cluster→customer cluster, the clear structure was architected. And on the base of the research on external mechanism, the rule and operational mode of the core internal mechanism was analyzed.2. The topological theory was adapted to research CRM customer cluster theory and to architect mathematics model. The concept: enterprise cluster, customer cluster, and information cluster, was put forward. Through the deep analysis on the intension of space, the family, base, subspace and reflection were illustrated.3. Four modes of customer relationship of CRM customer cluster mechanism were presented. According to the reflection principle of customer relationship, the corresponding analysis on the mode and sale entity of telecom enterprise was processed.4. The key link of CRM data mining, problems and algorithm on data pre-processing, was researched. The Out-lier algorithm based on density and attribution classical discrepant data protocol algorithm based on rough set theory were presented. The data pre-processing and data mining’s out-lier anomaly detection method of CRM were analyzed. And the pre-processing data reduction and algorithm adopted rough set theory were analyzed and programmed.5. Combining with the application of CRM system in provincial telecom enterprise, those theories were researched practically to prove the rightness of them and to get the更多还原