第五组第五组数据挖掘在客户流失管理中的应用 文献简介及选择理由12文献主要内容文献主要内容3结论结论4文献结构文献结构 第五组第五组文献简介及选择理由vData Mining Applications in Customer Churn Management 是一篇关于数据挖掘在顾客关系管理（客户流失管理）中的应用的文献综述，主要涉及各种数据挖掘技术和各种技术应用的统计。v客户流失管理是客户管理的核心，基于此研究背景贴近生活，易于理解。文中提到的多种数据挖掘方法模型覆盖面广，拓宽学习面。第五组第五组文献结构v引言：客户流失管理的重要性v介绍数据挖掘技术，方法分类v统计各种技术应用，研究趋势v结论、后续研究方向第五组第五组文献主要内容v顾客流失管理的重要性核心市场策略：保持现有客户防止客户流失 获得一个新客户的费用是留住一个客户费用的5-10倍在很多产业客户粘性增加5%即会带来25%-95%的净利润增加v研究背景及现状关于数据挖掘在客户流失管理的应用的文献综述很少现有文献，E.W.T. Ngai, Li Xiu and D.C.K. Chau, “Application of data mining techniques in customer relationship management: A Literature review and classification,”是对82篇文章的文献综述，从客户关系维度（客户识别、客户吸引、留住和发展客户）和数据挖掘分类（联合、分类、聚类、预测、回归分析、序列分析、可视化）进行研究。 欠缺：研究重点为客户粘性、客户流失管理，没有涉及具体的数据挖掘方法，分类不明确。第五组第五组文献主要内容 本文从数据挖掘方法出发对32篇文献进行分类，统计，研究趋势，填充了客户流失管理技术方法的空白v数据挖掘方法基本定义：神经网络或称作连接模型（ConnectionistModel） 它是一种模范动物神经网络行为特征，进行分布式并行信息处理的算法数学模型。这种网络依靠系统的复杂程度，通过调整内部大量节点之间相互连接的关系，从而达到处理信息的目的。第五组第五组数据挖掘方法基本定义v决策树（Decision Trees）一般都是自上而下的来生成的。每个决策或事件（即自然状态）都可能引出两个或多个事件，导致不同的结果。v回归分析（Logistic Regression)是确定两种或两种以上变数间相互依赖的定量关系的一种统计分析方法。v随机森林（Random Forests）是一个包含多个决策树的分类器， 并且其输出的类别是由个别树输出的类别的众数而定。可以产生高准确度的分类器，处理大量的输入变量。第五组第五组数据挖掘方法基本定义v支持向量机方法（Support Vector Machine）是分类的一个机器学习的过程。这是一种建立在统计学习理论的VC 维理论（对一个指标函数集，如果存在h个样本能够被函数集中的函数按所有可能的2h种形式分开，则称函数集能够把h个样本打散；函数集的VC维就是它能打散的最大样本数目h。）和结构风险最小原理基础上的，根据有限的样本信息在模型的复杂性（即对特定训练样本的学习精度）和学习能力（即无错误地识别任意样本的能力）之间寻求最佳折衷，以期获得最好的推广能力 。第五组第五组数据挖掘技术角度（Data Mining Perspective）对近期（03-09年）的32篇reference文献根据所采用的技术方法进行分类，分类如表一table1 对使用最多的前三种方法：神经网络（Neural Networks ）、决策树（Decision Trees）、回归分析（Logistic Regression）学习，作简要介绍。第五组第五组根据数据挖掘方法对32篇文献进行分类表一 给出各方法的使用频数，对于使用最多的前九种方法，选择在较符合各种指标的文献（表中黑体文献），作出简要解释。第五组第五组数据挖掘方法v神经网络（Neural Networks ）【17】P. C. Pendharkar, “Genetic algorithm based neural network approaches for predicting churn in cellular wireless network services,”使用遗传算法优化网络权重。研究基于移动通信环境，该文使用交叉验证（最大似然法cross-validation method）和精确模拟得出两个功能良好模型。该篇文献采用循环估计的方法，这一方法相对统计方法是一大改进。第五组第五组数据挖掘方法v决策树（Decision Tree ） 【11】 J. Qi et al., “ADTreesLogit model for customer churn prediction,”根据感受性曲线的单一指标（ROC曲线上各点反映着相同的感受性，它们都是对同一信号刺激的反应，是在几种不同的判定标准下所得的结果）选择输入变量，再将此变量分组，每个分组作为独立选择决策树模型的输入变量，得出的结果再作为洛基模型（logit model-一种回顾分析模型）的输入变量，使用回归特征消除的方法（Recursive feature elimination）忽略不相关因素，洛基模型将对客户流失进行预测。第五组第五组数据挖掘方法v回归分析（Logistic Regression） 【6】 Y. M. Zhang, J. Y. Qi, H. Y. Shu, and J. T. Cao, “A Hybrid KNN-LR Classifier and its Application in Customer Churn Prediction,”分析了独立性和目标参数之间的复杂关系，并采用K-近邻法（一种数据分类方法：将样本集中的每个样本都作为模板，用测试样本与每个模板做比较，看与哪 个模板最相似，就按最近似的模板的类别作为自己的类别）对这种复杂关系进行处理，找到依据单一的输入变量，改变每个独立的特征值。从准确性和感受曲线拟合上，这种方法在处理4个独立数据集时比较贴近。第五组第五组根据文献采用方法、发行年份维度对文献进行分类统计 表2： 神经网络一直是研究的重要方法。原因：在处理输入与输出的复杂非线性关系，抗噪（不受输入变量的不同类型干扰）时都十分适用。 自适应增量算法，梯度递增机制，线性判别分析等方法使用频率较低，还有待深入探究。第五组第五组分类统计柱状图第五组第五组出版物中客户流失管理（数据挖掘方法）文献比率 在这一研究领域的活跃出版物是Expert Systems with Applications ，感兴趣的同学可查阅相关文献，对此做深入研究。第五组第五组小结v本文从技术和统计角度对于数据挖掘在客户流失管理中的应用作出概述v给研究者和生产商提供这一领域的研究重点和趋势，并且出版商对于这一领域的文献越来越关注。v本文只涉及少量文献，对于知识挖掘，样本统计，模型评估未作深入探究。第五组第五组组员：尹鹏珍SA112040 杨金晶SA11204038 袁茜茜SA11204039 李璐涵SA11204043 赵 蕊SA11204040第五组第五组Referencesv1 E.W.T. Ngai, Li Xiu and D.C.K. Chau, “Application of data mining techniques in customer relationship management: A literature review and classification,” Expert Systems with Applications, vol. 36, 2009, pp. 25922602.v2 K. Coussement and Dirk Van den Poel, “Churn prediction in subscription services: An application of support vector machines while comparing two parameter-selection techniques,” Expert Systems with Applications, vol. 34, 2008, pp. 313327.v3 Bong-Horng Chu, Ming-Shian Tsai and Cheng-Seen Ho, “Toward a hybrid data mining model for customer retention,” Knowledge-Based Systems, vol. 20, 2007, pp. 703718.v4 X. Hu, “A Data Mining Approach for Retailing Bank Customer Attrition Analysis,” Applied Intelligence, vol. 22, 2005, pp. 4760, Springer.v5 H. S. Song, J. K. Kim, Y. B. Cho and S. H. Kim, “A Personalized Defection Detection and Prevention Procedure based on the Self-Organizing Map and Association Rule Mining: Applied to Online Game Site,” Artificial Intelligence Review, vol. 21, 2004, pp. 161184.v6 Y. M. Zhang, J. Y. Qi, H. Y. Shu, and J. T. Cao, “A Hybrid KNN-LR Classifier and its Application in Customer Churn Prediction,” Proc. the IEEE International Conference on Systems, Man and Cybernetics, Oct. 2007, pp. 32653269.v7 G. Song, D. Yang, L. Wu, T. Wang, Sh. Tang, “A Mixed Process Neural Network and its Application to Churn Prediction in Mobile Communications,” Proc. Sixth IEEE International Conference on Data Mining - Workshops (ICDMW06), 2006.v8 James J.H. Liou, “A novel decision rules approach for customer relationship management of the airline market,” Expert Systems with Applications, vol. 36 (3), April 2009, pp. 4374-4381.v9 M. Zan, Z. Shan, L. Li, L. Ai-jun, “A Predictive Model of Churn in Telecommunications Based on Data Mining,” Proc. IEEE International Conference on Control and Automation, IEEE Press, 2007.v10 Yi-Fan Wang, Ding-An Chiang, Mei-Hua Hsu, Cheng-Jung Lin, Ilong Lin, “A recommender system to avoid customer churn: A case study,” Expert Systems with Applications, vol. 36, 2009, pp. 8071 8075.第五组第五组Referencesv11 J. Qi et al., “ADTreesLogit model for customer churn prediction,” Annuls of Operations Research, vol. 168, 2009, pp. 247265, Springer.v12 Shin-Yuan Hung, David C. Yen and Hsiu-Yu Wang, “Applying data mining to telecom churn management,” Expert Systems with Applications, vol. 31, 2006, pp. 515524.v13 J. Zhaoa and Xing-Hua Dang, “Bank Customer Churn Prediction Based on Support Vector Machine: Taking a Commercial Banks VIP Customer Churn as the Example,” Proc. 4th International Conference on Wireless Communications, Networking and Mobile Computing, 2008 (WiCOM08), Oct. 2008, pp. 1-4.v14 Y. Xie and X. Li, “Churn Prediction with Linear Discriminant Boosting Algorithm,” Proc. the Seventh International Conference on Machine Learning and Cybernetics, Kunming, July 2008.v15 Hongmei Shao, Gaofeng Zheng and Fengxian An, “Construction of Bayesian Classifiers with GA for Predicting Customer Retention,” Proc. Fourth International Conference on Natural Computation, IEEE Computer Society Press, 2008.v16 Y. Xie, X. Li, E.W.T. Ngai and W. Ying, “Customer churn prediction using improved balanced random forests,” Expert Systems with Applications, vol. 36, 2009, pp. 54455449.v17 P. C. Pendharkar, “Genetic algorithm based neural network approaches for predicting churn in cellular wireless network services,” Expert Systems with Applications, vol. 36, 2009, pp. 6714- 6720.v18 K. Coussement, Dirk Van den Poel, “Improving customer attrition prediction by integrating emotions from client/company interaction emails and evaluating multiple classifiers,” Expert Systems with Applications, vol. 36, 2009, pp. 61276134.v19 S. Lessmann and S. Vo, “A reference model for customer-centric data mining with support vector machines,” European Journal of Operational Research, vol. 199 (2), Dec. 2009, pp. 520-530.第五组第五组Referencesv20 J. Burez and D. Van den Poel, “Handling class imbalance in customer churn prediction,” Expert Systems with Applications, vol. 36, 2009, 46264636.v21 Ding-An Chiang, Yi-Fan Wang, Shao-Lun Lee and Cheng-Jung Lin, “Goal-oriented sequential pattern for network banking churn analysis,” Expert Systems with Applications, vol. 25, 2003, pp. 293 302.v22 G. Zhang, “Customer Retention Based on BP ANN and Survival Analysis,” Proc. International Conference on Wireless Communications, Networking and Mobile Computing, 2007 (WiCom), Sept. 2007, pp. 3406-3411.v23 W. Buckinx and D. Van den Poel, “Customer base analysis: partial defection of behaviourally loyal clients in a non-contractual FMCG retail setting,” European Journal of Operational Research, vol. 164, 2005, pp. 252268.v24 B. Lariviere, D. Van den Poel, “Investigating the role of product features in preventing customer churn, by using survival analysis and choice modeling: The case of financial services,” Expert Systems with Applications, vol. 27, 2004, pp. 277285.v25 Lian Yan, Michael Fassino and Patrick Baldasare, “Predicting Customer Behavior via Calling Links,” Proc. International Joint Conference on Neural Networks, Montreal, Canada, August 2005.v26 E Xu, S. Liangshan, G. Xuedong and Z. Baofeng, “An Algorithm for Predicting Customer Churn via BP Neural Network Based on Rough Set,” Proc. the 2006 IEEE Asia-Pacific Conference on Services Computing (APSCC06).v27 “Predicting Customer Behavior in Telecommunications,” L. Yan, R.H. Wolniewicz, R. Dodier, IEEE Intelligent Systems, IEEE Computer Society.v28 N. Glady, B. Baesens and C. Croux, “Modeling churn using customer lifetime value,” European Journal of Operational Research, vol. 197, 2009, pp. 402411.v29 Jae-Hyeon Ahna, Sang-Pil Hana and Yung-Seop Lee, “Customer churn analysis: Churn determinants and mediation effects of partial defection in the Korean mobile telecommunications service industry,” Telecommunications Policy, vol. 30, 2006, pp. 552568.v30 K. Coussement and D. Van den Poel, “Integrating the voice of customers through call center emails into a decision support system for churn prediction,” Information & Management, vol. 45, 2008, pp. 164174.v31 B. Lariviere and D. Van den Poel, “Predicting customer retention and profitability by using random forests and regression forests techniques,” Expert Systems with Applications, vol. 29, 2005, pp. 472484.v32 J. Burez and D. Van den Poel, “Separating financial from commercial customer churn: A modeling step towards resolving the conflict between the sales and credit department,” Expert Systems with Applications, vol. 35, 2008, pp. 497514.v33 A. Prinzie T and D. Van den Poel, “Incorporating sequential information into traditional classification models by using an element/position-sensitive SAM,” Decision Support Systems, vol. 42, 2006, pp. 508526.