决策树和MBR在分析型CRM中的应用

探讨了数据挖掘技术在分析型CRM中的一个最典型的应用——如何寻找潜在客户。提出了一种寻找潜在客户的思路，基于这个思路，分别讨论了如何使用决策树和基于存储的推理（MBR）这两种数据挖掘技术来建立潜在客户预测模型，并给出了具体实例。     

一种基于数据挖掘的通用CRM系统框架及关键技术研究

文章在分析当前CRM的应用现状的基础上，综合数据挖掘技术、OLAP技术、知识库的最新研究成果．结合CRM的功能需求，研究了一种新的基于数据挖掘技术的通用商业CRM系统。该系统不仅能够提高用户对客户信息管理的效率，优化企业对客户的响应模式，同时可以根据行业特点进行快速定制实施．极大地方便了面向不同商业应用的客户关系管理平台的搭建，在实际商业应用中具有重要意义。     

客户关系管理系统与呼叫中心

客户关系管理（CRM）系统可以泛指企业获得和维持可带来业务收益的用户群的各种技术平台。随着市场竞争的日益激烈，企业迫切需要这样的技术平台，从而可以以更好的服务和支持（而不仅仅是产品特色）来赢得用户。几乎所有企业都需要CRM系统。因此，当CRM的概念被正式提出来以后，各类不同性质不同侧重点的解决方案都被冠以CRM的名头。其实，在现代商务中，CRM经常是从发展和管理与客户的沟通交流及各种关系开始的。目前，大多数国家普遍采用的处理这方面要求的技术平台就是呼叫中心（CallCenter）。1呼叫中心的概念呼叫中心是一组坐…     

决策树算法在银行信用评估中的研究

在贷款中尽可能地准确评估,提出了利用决策树理论对用户基本情况分析,建立了银行贷款客户信用评估中的决策树理想模型。为防止过拟合的问题出现,笔者对于最初生成的决策树又进行了裁剪修正,让决策模型在信用度决策获正确率较高,在不同测试集和验证集上取得了预期的效果。     

客户关系管理创新企业价值

本通过对客户关系管理(Customer Relationsgip Management缩写：CRM)的内含、流程、目的以及SWOT分析，阐明了在日趋激烈的市场竞争中，树立以客户为中心的经营理念，挖掘现有客户潜盈，注重企业价值链中上下游关系，借助并发挥CRM高效营销和协作组织优势，是电信运营商提升企业核心竞争力，创新企业价值的基础于核心。另就构建电信CRM系统，提出了具体建议。     

分析型CRM零售业不可缺少的软件工具

21世纪以来，中国经济迅速的发展使得企业竞争日益激烈，各行各业都在秣马厉兵，采取各种不同的销售方法争取更多的客户。对于零售业来说，直接面对的就是客户，如何降低成本有的放矢的销售、如何在吸引新客户的基础上保留老客户忠诚度、如何更新产品以满足客户日益增长的需求、如何推陈出新击败竞争对手如出一辙的销售模式，这些所面临的竞争问题更是毋庸置疑地显现在眼前。而CRM（客户关系管理）系统的产生无疑给出了答案，有了CRM系统，可以对客户的数据进行管理和划分，细分出各种类型，有效的组织企业资源，培养以客户为中心的经营行为以及实施以客户为中心的业务流程和销售模式，采取有针对性销售模式来降低成本，采取人性化的沟通促销方式来保留客户的忠诚度，采取销售数据分析法来确定大众用户的需求趋向，率先推出新产品，采取温馨关爱的方式正真想用户所想来满足用户的需求，而非机械的打折促销。通过与众不同的方式方法来提高企业的获利能力、收入以及客户满意度。最终取得客户满意，企业收益的双赢效果。     

呼叫中心与CRM的集成思路

CRM是一种改善企业与客户之间关系的新型管理机制，它贯穿企业的市场营销、销售、服务与技术支持等与客户相关的领域。其目的，一方面是通过提供更快和更周到的优质服务吸引和保持更多的客户；另一方面通过对业务流程的全面管理、实现商业规则的自动化以降低企业的经营成本。CRM涉及的核心软件包括Front office 、Back office 、BI三部分，目前绝大多数CRM厂商都偏重于其中的一部分。从企业的角度看，建设自己的CRM系统最常见的方法是集成多家供应商的产品，呼叫中心作为与客户接触的主要通道在其中起着关键作用。CRM首先是一种思想…     

电信CRM体系中的新型营销模式

CRM是英文Customer Relationship Management的缩写．中文译为“客户关系管理”,简单地说，CRM是一个获取、保持和增加可获利客户的过程，一种以客户为中心的商业模式。CRM涵盖了企业的营销、销售、服务等与客户接触的相关领域．它要求企业把客户当作企业运作的核心来组织自己的生产和服务。传统的企业是根据自己然后再把产品推销给客户的。     

基于数据挖掘技术的电子商务CRM研究

网络时代,电子商务CRM中存储海量客户数据,可利用数据挖掘技术对这些数据进行有效挖掘,发现有价值的信息。通过了解客户关系管理CRM的功能,分析数据挖掘技术的模式及过程,得出在电子商务CRM中可利用数据挖掘技术的分类模式获取新客户,聚类模式留住老客户,关联、序列模式提升客户价值,数据挖掘技术将在CRM中发挥越来越重要的作用。     

客户服务中心与CRM建设思路的探讨

CRM(客户关系管理)是以客户关系为中心的经营思想,用以组织化地实现管理客户关系的工作,树立良好的电信企业形象,提高客户满意度,从而使电信运营企业获得稳步增长的收益。CRM建设是当前各大企业关注的热点。分析了客户服务中心与CRM之间的关系,提出了以客户服务中心为核心的CRM实现方案,并指出了在CRM建设中需要规避的风险以及采取的措施。     

CRM在电信业的应用和实施

目前电信公司最大的需求是挖掘老客户的潜力,发展新客户,实施客户关系管理有助于企业解决这方面的问题。CRM项目的实施可分为3步:应用业务集成、业务数据分析和决策执行。成功实施CRM的关键因素包括确立合理可行的项目实施目标、高层管理者的理解和支持、让业务驱动CRM项目的实施、有效控制变更管理及选择软件供应商和实施伙伴。     

A Million Segments of One — How Personal Should Customer Relationship Management Get?

The foundation upon which customer relationship management (CRM) stands is a single, definitive view of customers which spans functions, channels, products and customer types and drives every customer interaction. CRM purports to recreate the ‘traditional corner shop’ experience to millions of clients. However, as techniques for data mining and pattern matching allow companies to build a relationship with customers, customers are questioning whether they actually want to have a relationship with companies. As CRM becomes more personalised, is it providing the experience that customers want? Are companies spending on personalised CRM only to reap limited benefits? This paper considers the different ways that companies such as Capital One, First Direct, Amazon and Tesco are personalising a CRM experience and examines the resulting relationship that they develop with their customers. Ultimately, the future of personalised CRM is dependent on non-intrusive, mutually beneficial and cost-effective strategies delivered through the appropriate medium for the customer. This requires recognition that CRM is more than just deployment of technology. It is a fusion of strategy, process and technology. Companies need to consider what customers want from them and whether the personalised solution that is being proposed reaps financial benefits. Building a million segments of one may work for some companies but others need to consider what kind of relationship their customers want from them and what they (and their customers) can gain from the data that they gather.     

CRM在中国电信业中的应用

客户关系管理(CRM)是基于数据仓库的信息分析系统。它利用数据仓库和数据挖掘技术分析客户的需求,从而为企业确定市场方向,巩固和发展新老客户提供依据。中国加入WTO后,面对即将来临的电信市场大战,国内电信运营商必须调整经营观念,加快CRM的建设。     

基于大数据挖掘的终端换机模型

目前,移动终端已成为运营商维系用户、拓展市场的战略重心,提升移动终端销量、扩大终端规模是各运营商的工作重点。基于数据挖掘技术,从用户属性、终端使用信息、终端搜索访问信息等维度出发,挖掘海量用户行为数据价值,建立终端换机模型,具体包括基于决策树算法的用户换机倾向识别模型和基于聚类算法的终端推荐模型,助力移动终端精准营销。     

Research of CRM Based on Customer Intelligence Engine

The discipline of business intelligence addresses a broad range of functional activities from data mining and statistical analysis to predictive modeling and reporting, and customer intelligence is the actionable output from an intelligence eco-system. In order to focus enterprise's attention on their customers satisfaction in the customer relationship management and make CRM system run more efficiently, a new concept of customer intelligence engine(CIE) is proposed at first time in the paper, the architecture of CIE is structured, the trigger of CIE is defined and described, the CIE-based CRM eco-system is also discussed.     

基于决策树的网络客户分类方法研究

传统的网络购物只是对商品进行一个简单的分类和陈列,对于电子商务的商家并没有对网络消费者的购物数据进行深入研究探讨.针对网络购物过程中消费者选择商品的趋向性的不同,引入了基于决策树分类方法对网络客户购买商品的行为进行分析,并从决策树中挖掘出影响网络购物的主要因素以及各因素对网络购买行为的强弱影响程度.实验结果表明,此方法可以有效的对网络客户进行分类,有利于决策分析.     

基于数据挖掘的电信客户流失预测分析

针对电信客户日益严重地流失问题,通过某电信运营商的历史资料,对电信PAS流失客户的自然属性和行为属性进行研究,利用决策树算法建立了客户流失预测模型。通过对模型进行评估分析,得到预测效果较好的模型,最后加入成本因素,进一步优化了模型。     

电力潜在敏感客户预测的大数据方法应用

随着 95598 业务的不断发展延伸,人工话务强度增大。为了进一步加深对客户隐性特征以及诉求的认识和理解,提升 95598 人工精细化客户服务水平,对投诉倾向等客户服务中的典型应用场景进行了需求细化。基于电力服务工单数据,选取建模关键指标,通过熵权法、主成分分析和决策树等数据挖掘算法,对潜在投诉倾向客户和计划停电敏感客户进行识别,以便有针对性地进行服务资源调度,充分做好应对措施,有效减少投诉压力,提升服务精度。     

The associate impact of individual internal experiences and reference groups on buying behavior: A case study of animations,comics, and games consumers

Among the entertainment and media market, it can be observed that animations, comics, and video games (hereinafter abbreviated as ACG) have the highest output value and most market influence. Moreover, ACG also incorporates various industries and creates many derivative products. As the ACG industry emphasizes acousto-optics, imagery, and storylines, personal impressions derived from consumer experiences will influence consumer decisions. In addition, the ACG industry is mainly marketed towards younger age groups, with younger people being the main consumers; as such, these consumers’ decisions are more easily affected by peer behavior.This study aims to analyze the effects of internal cognitions and external influences on buying behavior of ACG consumers by applying the uncomplicated decision tree data mining algorithm. We analyze and develop the target attributes on measures of customer loyalty for ACG industry to set up the decision trees from the collected questionnaire data. The decision tree data mining method is applied to analyze the hidden association rules between the target attributes (i.e., consumer loyalty) and the critical influencing factors of consumer’s internal impressions and external influences for ACG consumers. The results and suggestions of this paper can be used as a reference for enterprises in the ACG industry to help with business policies concerning products’ extensional design, marketing, and CRM, and to further strengthen customer satisfaction and loyalty, thus increasing company profits.     

C-fuzzy decision trees

This paper introduces a concept and design of decision trees based on information granules - multivariable entities characterized by high homogeneity (low variability). As such granules are developed via fuzzy clustering and play a pivotal role in the growth of the decision trees, they will be referred to as C-fuzzy decision trees. In contrast with "standard" decision trees in which one variable (feature) is considered at a time, this form of decision trees involves all variables that are considered at each node of the tree. Obviously, this gives rise to a completely new geometry of the partition of the feature space that is quite different from the guillotine cuts implemented by standard decision trees. The growth of the C-decision tree is realized by expanding a node of tree characterized by the highest variability of the information granule residing there. This paper shows how the tree is grown depending on some additional node expansion criteria such as cardinality (number of data) at a given node and a level of structural dependencies (structurability) of data existing there. A series of experiments is reported using both synthetic and machine learning data sets. The results are compared with those produced by the "standard" version of the decision tree (namely, C4.5).