Predicting time-to-churn of prepaid mobile telephone customers using social network analysis

Mobile phone carriers in a saturated market must focus on customer retention to maintain profitability. This study investigates the incorporation of social network information into churn prediction models to improve accuracy, timeliness, and profitability. Traditional models are built using customer attributes, however these data are often incomplete for prepaid customers. Alternatively, call record graphs that are current and complete for all customers can be analysed. A procedure was developed to build the call graph and extract relevant features from it to be used in classification models. The scalability and applicability of this technique are demonstrated on a telecommunications data set containing 1.4 million customers and over 30 million calls each month. The models are evaluated based on ROC plots, lift curves, and expected profitability. The results show how using network features can improve performance over local features while retaining high interpretability and usability.     

New insights into churn prediction in the telecommunication sector: A profit driven data mining approach

Customer churn prediction models aim to indicate the customers with the highest propensity to attrite, allowing to improve the efficiency of customer retention campaigns and to reduce the costs associated with churn. Although cost reduction is their prime objective, churn prediction models are typically evaluated using statistically based performance measures, resulting in suboptimal model selection. Therefore, in the first part of this paper, a novel, profit centric performance measure is developed, by calculating the maximum profit that can be generated by including the optimal fraction of customers with the highest predicted probabilities to attrite in a retention campaign. The novel measure selects the optimal model and fraction of customers to include, yielding a significant increase in profits compared to statistical measures.In the second part an extensive benchmarking experiment is conducted, evaluating various classification techniques applied on eleven real-life data sets from telecom operators worldwide by using both the profit centric and statistically based performance measures. The experimental results show that a small number of variables suffices to predict churn with high accuracy, and that oversampling generally does not improve the performance significantly. Finally, a large group of classifiers is found to yield comparable performance.     

Multicriteria evidential reasoning decision modelling and analysis—prioritizing voices of customer

In this paper, a new methodology is investigated to support the prioritization of the voices of customers through various customer satisfaction surveys. This new methodology consists of two key components: an innovative evidence-driven decision modelling framework for representing and transforming large amounts of data sets and a generic reasoning-based decision support process for aggregating evidence to prioritize the voices of customer on the basis of the Evidential Reasoning (ER) approach. Methods and frameworks for data collection and representation via multiple customer satisfaction surveys were examined first and the distinctive features of quantitative and qualitative survey data are analysed. Several novel yet natural and pragmatic rule-based functions are then proposed to transform survey data systematically and consistently from different measurement scales to a common scale, with the original features and profiles of the data preserved in the transformation process. These new transformation functions are proposed to mimic expert judgement processes and designed to be sufficiently flexible and rigorous so that expert judgements and domain specific knowledge can be taken into account naturally, systematically and consistently in the transformation process. The ER approach is used for synthesizing quantitative and qualitative data under uncertainty that can be caused due to missing data and ambiguous survey questions. A new generic method is also proposed for ranking the voices of customer based on qualitative measurement scales without having to quantify assessment grades to fixed numerical values. A case study is examined using an Intelligent Decision System (IDS) to illustrate the application of the decision modelling framework and decision support process for prioritizing the voices of customer for a world-leading car manufacturer.     

Predicting partial customer churn using Markov for discrimination for modeling first purchase sequences

Currently, in order to remain competitive companies are adopting customer centered strategies and consequently customer relationship management is gaining increasing importance. In this context, customer retention deserves particular attention. This paper proposes a model for partial churn detection in the retail grocery sector that includes as a predictor the similarity of the products?? first purchase sequence with churner and non-churner sequences. The sequence of first purchase events is modeled using Markov for discrimination. Two classification techniques are used in the empirical study: logistic regression and random forests. A real sample of approximately 95,000 new customers is analyzed taken from the data warehouse of a European retailing company. The empirical results reveal the relevance of the inclusion of a products?? sequence likelihood in partial churn prediction models, as well as the supremacy of logistic regression when compared with random forests.     

A methodology based on profitability criteria for defining the partial defection of customers in non-contractual settings

The defection or churn of customers represents an important concern for any company and a central matter of interest in customer base analysis. An additional complication arises in non-contractual settings, where the characteristics that should be observed to saying that a customer has totally or partially defected are not clearly defined. As a matter of fact, different definitions of the churn situation could be used in this context. Focusing on non-contractual settings, in this paper we propose a methodology for evaluating the short-time economic effects that using a certain definition of churn would have on a company. With this aim, we have defined two efficiency measures for the economic results of a marketing campaign implemented against churn, and these measures have been computed using a set of definitions of partial defection. Our methodology finds that definition maximizing both efficiency measures and moreover, the monetary amount that the company should invest per customer in the campaign for achieving the optimal solution. This has been modelled as a multiobjective optimization problem that we solved using compromise programming. Numerical results using real data from a Spanish retailing company are presented and discussed in order to show the performance and validity of our proposal.     

A hierarchical multiple kernel support vector machine for customer churn prediction using longitudinal behavioral data

The availability of abundant data posts a challenge to integrate static customer data and longitudinal behavioral data to improve performance in customer churn prediction. Usually, longitudinal behavioral data are transformed into static data before being included in a prediction model. In this study, a framework with ensemble techniques is presented for customer churn prediction directly using longitudinal behavioral data. A novel approach called the hierarchical multiple kernel support vector machine (H-MK-SVM) is formulated. A three phase training algorithm for the H-MK-SVM is developed, implemented and tested. The H-MK-SVM constructs a classification function by estimating the coefficients of both static and longitudinal behavioral variables in the training process without transformation of the longitudinal behavioral data. The training process of the H-MK-SVM is also a feature selection and time subsequence selection process because the sparse non-zero coefficients correspond to the variables selected. Computational experiments using three real-world databases were conducted. Computational results using multiple criteria measuring performance show that the H-MK-SVM directly using longitudinal behavioral data performs better than currently available classifiers.     

Comparison of some AI and statistical classification methods for a marketing case

Recent progress in data processing technology has made the accumulation and systematic organization of large volumes of data a routine activity. As a result of these developments, there is an increasing need for data-based or data-driven methods of model development. This paper describes data-driven classification methods and shows that the automatic development and refinement of decision support models is now possible when the machine is given a large (or sometimes even a small) amount of observations that express instances of a certain task domain. The classifier obtained may be used to build a decision support system, to refine or update an existing system and to understand or improve a decision-making process. The described AI classification methods are compared with statistical classification methods for a marketing application. They can act as a basis for data-driven decision support systems that have two basic components: an automated knowledge module and an advice module or, in different terms, an automated knowledge acquisition/retrieval module and a knowledge processing module. When these modules are integrated or linked, a decision support system can be created which enables an organization to make better-quality decisions, with reduced variance, probably using fewer people.     

Analyzing consumers’ shopping behavior using RFID data and pattern mining

The development of sensor networks has enabled detailed tracking of customer behavior in stores. Shopping path data which records each customer??s position and time information is attracting attention as new marketing data. However, there are no proposed marketing models which can identify good customers from huge amounts of time series data on customer movement in the store. This research aims to use shopping path data resulting from tracking customer behavior in the store, using information on the sequence of visiting each product zone in the store and staying time at each product zone, to find how they affect purchasing. To discover useful knowledge for store management, shopping paths data has been transformed into sequence data including information on visit sequence and staying times in the store, and LCMseq has been applied to them to extract frequent sequence patterns. In this paper, we find characteristic in-store behavior patterns of good customers by using actual data of a Japanese supermarket.     

On the customer lifetime value: a mathematical perspective

The customer lifetime value (CLV) is an important concept increasingly considered in the field of general marketing and in the management of firms, of organizations to increase the captured profitability. It represents the total value that a customer produces during his or her lifetime, or better represents the measure of the potential profit generating by a customer. The companies use the customer lifetime value to segment customers, analyze probability of churn, allocate resources or formulate strategies and, therefore, they increasingly derive revenue from the creation and from sustenance of long-term relationships with their customers. For this reason, the customer lifetime value is increasingly considered a touchstone for the management of customer relationships. In this article, the authors deepen the concept and use of customer lifetime value and present some mathematical models for its determination. There is many models for this purpose but most of them are theoretic, complex and not applicable. Though not exhaustive, the major contribution of this paper is that it provides a general mathematical formulation to estimate the CLV and that it has a context less specific compared to papers, present in literature, on the customer lifetime.     

ADTreesLogit model for customer churn prediction

In this paper, we propose ADTreesLogit, a model that integrates the advantage of ADTrees model and the logistic regression model, to improve the predictive accuracy and interpretability of existing churn prediction models. We show that the overall predictive accuracy of ADTreesLogit model compares favorably with that of TreeNet®, a model which won the Gold Prize in the 2003 mobile customer churn prediction modeling contest (The Duke/NCR Teradata Churn Modeling Tournament). In fact, ADTreesLogit has better predictive accuracy than TreeNet® on two important observation points.     

Selecting optimal selling format of a product in B2C online auctions with boundedly rational customers

The advancement of Internet technology has enabled new formats for selling products in the B2C online auctions. At present, on the major online auction sites, there exist three popular selling formats, namely, the posted price, pure auction and buy-price auction formats. It is an important decision problem for a firm to select the most profitable format to sell its products through the Internet. The customer behavior is of course a crucial element of the decision process. To the best of our knowledge, most models available today assume that customers are perfectly rational. To better understand the decision process, in this paper, we incorporate the concept of bounded rationality into consideration. We first present a “behavior choice function” to characterize the behavior of the customers with bounded rationality. Then corresponding to each selling format, we construct a revenue model based on the bounded rationality for analysis. Finally, we conduct some elaborate computational experiments to investigate the performance of each revenue model for developing new managerial insights. Our computational results clearly demonstrate how the bounded rationality of customer behavior affects the choice of a preferable selling format for a B2C firm in an online auction.     

Modeling churn using customer lifetime value

The definition and modeling of customer loyalty have been central issues in customer relationship management since many years. Recent papers propose solutions to detect customers that are becoming less loyal, also called churners. The churner status is then defined as a function of the volume of commercial transactions. In the context of a Belgian retail financial service company, our first contribution is to redefine the notion of customer loyalty by considering it from a customer-centric viewpoint instead of a product-centric one. We hereby use the customer lifetime value (CLV) defined as the discounted value of future marginal earnings, based on the customer’s activity. Hence, a churner is defined as someone whose CLV, thus the related marginal profit, is decreasing. As a second contribution, the loss incurred by the CLV decrease is used to appraise the cost to misclassify a customer by introducing a new loss function. In the empirical study, we compare the accuracy of various classification techniques commonly used in the domain of churn prediction, including two cost-sensitive classifiers. Our final conclusion is that since profit is what really matters in a commercial environment, standard statistical accuracy measures for prediction need to be revised and a more profit oriented focus may be desirable.     

Extending decision support systems: The integration of data,knowledge, and model management

The proliferation of desktop computing has once again rekindled the interest in making computerized tools available to managers and other decision makers. This paper elaborates on a model that integrates data, knowledge, and model management and shows how decision support systems (DSSs) can be extended to support managers in a truly novel way. The model, the Knowledge/Data Model (KDM), is explained and the significance of its applicability to the management of data, knowledge, and models is illustrated through several examples. KDM continues to evolve and is being applied to domains from computer chip design to production and inventory management systems.     

Kernel-based learning methods for preference aggregation

The mathematical representation of human preferences has been a subject of study for researchers in different fields. In multi-criteria decision making (MCDM) and fuzzy modeling, preference models are typically constructed by interacting with the human decision maker (DM). However, it is known that a DM often has difficulties to specify precise values for certain parameters of the model. He/she instead feels more comfortable to give holistic judgements for some of the alternatives. Inference and elicitation procedures then assist the DM to find a satisfactory model and to assess unjudged alternatives. In a related but more statistical way, machine learning algorithms can also infer preference models with similar setups and purposes, but here less interaction with the DM is required/allowed. In this article we discuss the main differences between both types of inference and, in particular, we present a hybrid approach that combines the best of both worlds. This approach consists of a very general kernel-based framework for constructing and inferring preference models. Additive models, for which interpretability is preserved, and utility models can be considered as special cases. Besides generality, important benefits of this approach are its robustness to noise and good scalability. We show in detail how this framework can be utilized to aggregate single-criterion outranking relations, resulting in a flexible class of preference models for which domain knowledge can be specified by a DM.      

移动电话客户流失数据挖掘

本文首先回顾了顾客流失的相关文献,然后利用统计分析方法和数据挖掘技术分析了移动电话号码与移动电话型号对客户流失的影响,对分析结果进行了解释,并给出一些营销建议。     

50 years of data mining and OR: upcoming trends and challenges

Data mining involves extracting interesting patterns from data and can be found at the heart of operational research (OR), as its aim is to create and enhance decision support systems. Even in the early days, some data mining approaches relied on traditional OR methods such as linear programming and forecasting, and modern data mining methods are based on a wide variety of OR methods including linear and quadratic optimization, genetic algorithms and concepts based on artificial ant colonies. The use of data mining has rapidly become widespread, with applications in domains ranging from credit risk, marketing, and fraud detection to counter-terrorism. In all of these, data mining is increasingly playing a key role in decision making. Nonetheless, many challenges still need to be tackled, ranging from data quality issues to the problem of how to include domain experts' knowledge, or how to monitor model performance. In this paper, we outline a series of upcoming trends and challenges for data mining and its role within OR.     

Preference disaggregation and statistical learning for multicriteria decision support: A review

Disaggregation methods have become popular in multicriteria decision aiding (MCDA) for eliciting preferential information and constructing decision models from decision examples. From a statistical point of view, data mining and machine learning are also involved with similar problems, mainly with regard to identifying patterns and extracting knowledge from data. Recent research has also focused on the introduction of specific domain knowledge in machine learning algorithms. Thus, the connections between disaggregation methods in MCDA and traditional machine learning tools are becoming stronger. In this paper the relationships between the two fields are explored. The differences and similarities between the two approaches are identified, and a review is given regarding the integration of the two fields.     

Object technology software selection: a case study

The goal of this paper is to present a case study focusing on object technology assessment in a computer service industry. We develop decision models to give proper recommendations for object-oriented software projects. The assessment uses a quantitative approach, in which a mixed-integer linear programming model and a multi-objective model were formulated and applied. By reducing the element of subjectivity, these formal models led to consistent tool selection. By separating the data and models, the models can be reused in subsequent software development projects. Finally, by allowing users to specify their objectives and requirements and by providing a sensitivity analysis of the results, this approach also increases customer orientation.     

用LDA Boosting算法进行客户流失预测

本文提出一种LDA boost(Linear Discriminant Analysis boost)分类方法,该算法能有效利用样本的所有特征,并且能够从高维特征空间里提取并组合优化出最具有判别能力的低维特征,使得样本类间离散度和类内离散度的比值最大,从而不会产生过度学习,大大提高算法效率。该算法有效性在某商业银行的客户流失预测过程的真实数据集中得到了验证。与其他同类算法,如人工神经网络、决策树、支持向量机等运算结果相比,该方法可以显著提高运算精度。同时,LDAboosting与其他boosting算法相比,也具有显著的优越性。     

不透明销售的顾客退货政策选择

本文探讨从事不透明销售的零售商对顾客退货政策的选择问题。分别针对零售商垄断和竞争两种市场情况,建立不透明零售商与其它供应链成员(制造商或普通零售商)之间的博弈模型,获得唯一均衡解;对均衡结果进行结构化分析,给出不透明销售方式下采用全额退款政策的判别条件;针对均衡结果,分析零售商垄断情况下产品不透明参数的最优设计,以及零售商竞争情况下的市场分化情况;鉴于净残值参数在退货政策选择中的决定性作用,本文进一步探讨了净残值为正时全额退款政策对各参与方利润及产品需求和价格的影响,分析了净残值在其中的作用机理。本研究能够为不透明零售商制定退货政策和价格以及其它供应链成员制定相关决策提供支持。