Wouter Verbeke

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.

Data Mining Techniques for Software Effort Estimation: A Comparative Study

A predictive model is required to be accurate and comprehensible in order to inspire confidence in a business setting. Both aspects have been assessed in a software effort estimation setting by previous studies. However, no univocal conclusion as to which technique is the most suited has been reached. This study addresses this issue by reporting on the results of a large scale benchmarking study. Different types of techniques are under consideration, including techniques inducing tree/rule-based models like M5 and CART, linear models such as various types of linear regression, nonlinear models (MARS, multilayered perceptron neural networks, radial basis function networks, and least squares support vector machines), and estimation techniques that do not explicitly induce a model (e.g., a case-based reasoning approach). Furthermore, the aspect of feature subset selection by using a generic backward input selection wrapper is investigated. The results are subjected to rigorous statistical testing and indicate that ordinary least squares regression in combination with a logarithmic transformation performs best. Another key finding is that by selecting a subset of highly predictive attributes such as project size, development, and environment related attributes, typically a significant increase in estimation accuracy can be obtained.

Social network analysis for customer churn prediction

This study examines the use of social network information for customer churn prediction. An alternative modeling approach using relational learning algorithms is developed to incorporate social network effects within a customer churn prediction setting, in order to handle large scale networks, a time dependent class label, and a skewed class distribution. An innovative approach to incorporate non-Markovian network effects within relational classifiers and a novel parallel modeling setup to combine a relational and non-relational classification model are introduced. The results of two real life case studies on large scale telco data sets are presented, containing both networked (call detail records) and non-networked (customer related) information about millions of subscribers. A significant impact of social network effects, including non-Markovian effects, on the performance of a customer churn prediction model is found, and the parallel model setup is shown to boost the profits generated by a retention campaign.

A Novel Profit Maximizing Metric for Measuring Classification Performance of Customer Churn Prediction Models

The interest for data mining techniques has increased tremendously during the past decades, and numerous classification techniques have been applied in a wide range of business applications. Hence, the need for adequate performance measures has become more important than ever. In this paper, a cost-benefit analysis framework is formalized in order to define performance measures which are aligned with the main objectives of the end users, i.e., profit maximization. A new performance measure is defined, the expected maximum profit criterion. This general framework is then applied to the customer churn problem with its particular cost-benefit structure. The advantage of this approach is that it assists companies with selecting the classifier which maximizes the profit. Moreover, it aids with the practical implementation in the sense that it provides guidance about the fraction of the customer base to be included in the retention campaign.

Conventional, Hybrid, or Electric Vehicles: Which Technology for an Urban Distribution Centre?

Freight transport has an important impact on urban welfare. It is estimated to be responsible for 25% of CO2 emissions and up to 50% of particles matters generated by the transport sector in cities. Facing that problem, the European Commission set the objective of reaching free CO2 city logistics by 2030 in major urban areas. In order to achieve this goal, electric vehicles could be an important part of the solution. However, this technology still faces a number of barriers, in particular high purchase costs and limited driving range. This paper explores the possible integration of electric vehicles in urban logistics operations. In order to answer this research question, the authors have developed a fleet size and mix vehicle routing problem with time windows for electric vehicles. In particular, an energy consumption model is integrated in order to consider variable range of electric vehicles. Based on generated instances, the authors analyse different sets of vehicles in terms of vehicle class (quadricycles, small vans, large vans, and trucks) and vehicle technology (petrol, hybrid, diesel, and electric vehicles). Results show that a fleet with different technologies has the opportunity of reducing costs of the last mile.

Profit optimizing customer churn prediction with Bayesian network classifiers

Customer churn prediction is becoming an increasingly important business analytics problem for telecom operators. In order to increase the efficiency of customer retention campaigns, churn prediction models need to be accurate as well as compact and interpretable. Although a myriad of techniques for churn prediction has been examined, there has been little attention for the use of Bayesian Network classifiers. This paper investigates the predictive power of a number of Bayesian Network algorithms, ranging from the Naive Bayes classifier to General Bayesian Network classifiers. Furthermore, a feature selection method based on the concept of the Markov Blanket, which is genuinely related to Bayesian Networks, is tested. The performance of the classifiers is evaluated with both the Area under the Receiver Operating Characteristic Curve and the recently introduced Maximum Profit criterion. The Maximum Profit criterion performs an intelligent optimization by targeting this fraction of the customer base which would maximize the profit generated by a retention campaign. The results of the experiments are rigorously tested and indicate that most of the analyzed techniques have a comparable performance. Some methods, however, are more preferred since they lead to compact networks, which enhances the interpretability and comprehensibility of the churn prediction models.

Social network analytics for churn prediction in telco

Comparison of Social Networks Analytics methods for predicting churn in telco.Ranking of 24 relational learners with respect to predictive performance.Collective inferencing does not improve the performance of relational classifiers.The best models apply a classifier with network features and relational learner scores.Network construction matters for model performance. Social network analytics methods are being used in the telecommunication industry to predict customer churn with great success. In particular it has been shown that relational learners adapted to this specific problem enhance the performance of predictive models. In the current study we benchmark different strategies for constructing a relational learner by applying them to a total of eight distinct call-detail record datasets, originating from telecommunication organizations across the world. We statistically evaluate the effect of relational classifiers and collective inference methods on the predictive power of relational learners, as well as the performance of models where relational learners are combined with traditional methods of predicting customer churn in the telecommunication industry. Finally we investigate the effect of network construction on model performance; our findings imply that the definition of edges and weights in the network does have an impact on the results of the predictive models. As a result of the study, the best configuration is a non-relational learner enriched with network variables, without collective inference, using binary weights and undirected networks. In addition, we provide guidelines on how to apply social networks analytics for churn prediction in the telecommunication industry in an optimal way, ranging from network architecture to model building and evaluation.

A comparative study of social network classifiers for predicting churn in the telecommunication industry

Relational learning in networked data has been shown to be effective in a number of studies. Relational learners, composed of relational classifiers and collective inference methods, enable the inference of nodes in a network given the existence and strength of links to other nodes. These methods have been adapted to predict customer churn in telecommunication companies showing that incorporating them may give more accurate predictions. In this research, the performance of a variety of relational learners is compared by applying them to a number of CDR datasets originating from the telecommunication industry, with the goal to rank them as a whole and investigate the effects of relational classifiers and collective inference methods separately. Our results show that collective inference methods do not improve the performance of relational classifiers and the best performing relational classifier is the network-only link-based classifier, which builds a logistic model using link-based measures for the nodes in the network.

Using Social Network Classifiers for Predicting E-Commerce Adoption

This paper indicates that knowledge about a person’s social network is valuable to predict the intent to purchase books and computers online. Data was gathered about a network of 681 persons and their intent to buy products online. Results of a range of networked classification techniques are compared with the predictive power of logistic regression. This comparison indicates that information about a person’s social network is more valuable to predict a person’s intent to buy online than the person’s characteristics such as age, gender, his intensity of computer use and his enjoyment when working with the computer.

Recommendation-Based Conceptual Modeling and Ontology Evolution Framework (CMOE+)

Within an enterprise, various stakeholders create different conceptual models, such as process, data, and requirements models. These models are fundamentally based on similar underlying enterprise (domain) concepts, but they differ in focus, use different modeling languages, take different viewpoints, utilize different terminology, and are used to develop different enterprise artifacts; as such, they typically lack consistency and interoperability. This issue can be solved by enterprise-specific ontologies, which serve as a reference during the conceptual model creation. Using such a shared semantic repository makes conceptual models interoperable and facilitates model integration. The challenge to accomplish this is twofold: on the one hand, an up-to-date enterprise-specific ontology needs to be created and maintained, and on the other hand, different modelers also need to be supported in their use of the enterprise-specific ontology. The authors propose to tackle these challenges by means of a recommendation-based conceptual modeling and an ontology evolution framework, and we focus in particular on ontology-based modeling support. To this end, the authors present a framework for Business Process Modeling Notation (BPMN) as a conceptual modeling language, and focus on how modelers can be assisted during the modeling process and how this impacts the semantic quality of the resulting models. Subsequently, a first, large-scale explorative experiment is presented involving 140 business students to evaluate the BPMN instantiation of our framework. The experiments show promising results with regard to incurred overheads, intention of use and model interoperability.