Predicting subscriber dissatisfaction and improving retention in the wireless telecommunications industry

Mozer, Michael C.; Wolniewicz, Richard; Grimes, David B.; Johnson, Eric L.; Kaushansky, Howard

doi:10.1109/72.846740

Cited by 287 publications

(160 citation statements)

References 4 publications

Supporting

Mentioning

148

Contrasting

Unclassified

Order By: Relevance

“…For this approach, we describe some of the most referred works relevant to the proposed method. A pioneering work in this area was published by Mozer et al (2000), where the authors explored techniques from statistical machine learning to predict churn and subsequently determine what incentives should be offered to subscribers to improve retention and maximize profitability to the carrier. The techniques include (1) logistic regression, (2) the use of decision trees, (3) the use of neural networks, (4) the boosting of decision trees with the AdaBoost algorithm, and (5) the boosting of neural networks with the AdaBoost algorithm.…”

Section: Churn Prediction In the Literaturementioning

confidence: 99%

A diffusion model for churn prediction based on sociometric theory

Droftina

Štular

Košir

2014

Adv Data Anal Classif

View full text Add to dashboard Cite

Churn prediction has received much attention in the last decade. With the evolution of social networks and social network analysis tools in recent years, the consideration of social ties in churn prediction has proven promising. One possibility is to use energy diffusion models to model the spread of influence through a social network. This paper proposes a novel churn prediction diffusion model based on sociometric clique and social status theory. It describes the concept of energy in the diffusion model as an opinion of users, which is transformed to user influence using the derived social status function. Furthermore, a novel diffusion model prediction scheme applicable to a single user or a small subset of users is described: the Targeted User Subset Churn Prediction Scheme. The scheme allows fast churn prediction using limited computing resources. The diffusion model is evaluated on a real dataset of users obtained from the largest Slovenian mobile service provider, using the F-measure and lift curve. The empirical results show a significant improvement in prediction accuracy of the proposed method compared with the basic spreading activation technique (SPA) diffusion model. More specifically, our approach outperforms a basic SPA diffusion model by 116 % in terms of lift in the fifth percentile.U. Droftina (B) · M. Štular

show abstract

Section: Churn Prediction In the Literaturementioning

confidence: 99%

A diffusion model for churn prediction based on sociometric theory

Droftina

Štular

Košir

2014

Adv Data Anal Classif

View full text Add to dashboard Cite

show abstract

“…Other studies [5], [6], [7], [8] , [9], [10], [11], [12], [13] proposed different approaches for churn prediction in which users were considered individually.…”

Section: Previous Researchmentioning

confidence: 99%

“…In [11], Mozer et al experimented with various prediction models, such as logistic regression, decision trees, and neural networks to predict churn among mobile phone users.…”

Section: Previous Researchmentioning

confidence: 99%

Impact of Social Network to Churn in Mobile Network

2015

View full text Add to dashboard Cite

Original scientific paperAs the telecommunications sector has reached its mature stage, maintaining existing users has become crucial for service providers. Analyzing the call data records, it is possible to observe their users in the context of social network and obtain additional insights about the spread of influence among interconnected users, which is relevant to churn. In this paper, we examine the communication patterns of mobile phone users and subscription plan logs. Our goal is to use a simple model to predict which users are most likely to churn, solely by observing each user's social network, which is formed by outgoing calls, and churn among their neighbours. To measure the importance of social network parameters with regard to churn prediction, we compare three models: spatial classification, regression model, and artificial neural networks. For each subscriber, we observe three social network parameters, the number of neighbors that have churned, the number of calls to these neighbors, and the duration of these calls for different time periods. The results indicate that using only one or two of these parameters yields results that are comparable or better than the complex models with large amounts of individual and/or social network input parameters that other researchers have proposed.Key words: Churn, Social Network Analysis, Machine Learning, Mobile Network Utjecaj društvene mreže na odljev korisnika u mobilnim mrežama. Kako je telekomunikacijski sektor dosegao zreli stadij, zadržavanje postojećih korisnika od ključne je važnosti za pružatelje telekomunikacijskih usluga. Analizom liste poziva moguće je nadzirati korisnike u kontekstu društvene mreže i dobiti dodatni uvid u širenje utjecaja među povezanim korisnicima, što je relevantno za odljev korisnika. U ovom radu razmatramo obrasce komunikacije korisnika mobilnih mreža i podatke o planu pretplate. Naš cilj je korištenjem jednostavnog modela predvidjeti koji korisnici su najskloniji prijelazu na drugu mrežu, pritom koristeći samo korisnikovu društvenu mrežu koja se formira odlaznim pozivima i prijelazima između mreža njihovih susjeda. S ciljem mjerenja važnosti pojedinog parametra društvene mreže za predikciju prelaska na drugu mrežu uspoređena su tri modela: prostorna klasifikacija, regresijski model i model neuronske mreže. Za svakog pretplatnika razmatramo tri parametra društvene mreže: broj susjeda koji su promijenili mrežu, broj poziva prema njima kao i trajanje spomenutih poziva u različitim vremenskim razdobljima. Rezultati pokazuju kako se korištenjem samo jednog ili dva od navedenih parametara društvene mreže postižu rezultati koji su usporedivi ili bolji od rezultata složenijih modela drugih autora koji koriste veliki broj osobnih parametara i/ili parametara društvene mreže.

show abstract

“…• Operator 1: This telecommunication dataset was originally studied by [28], and contains data from 47,761 customers described by 47 variables. It was used for benchmarking machine learning methods in [35] under the name of Operator 1 (O1).…”

Section: Description Of Datasetsmentioning

confidence: 99%

Profit-based feature selection using support vector machines – General framework and an application for customer retention

Maldonado

Flores

Verbraken

et al. 2015

Applied Soft Computing

View full text Add to dashboard Cite

A c c e p t e d M a n u s c r i p t  A novel profit-based feature selection method for churn prediction with SVM is presented.  A backward elimination algorithm is performed to maximize the profit of a retention campaign. Our experiments on churn prediction datasets underline the potential of the proposed approaches. *Highlights (for review)Page 2 of 36 A c c e p t e d M a n u s c r i p Effective churners Outflow New customers InflowConsists of N customers with average customer lifetime value CLV.The cost of contacting a customer is f.The cost of an incentive offer is dWould-be churnersClassified as churners (η η η η)Would-be churnersClassified as nonchurners (1-η η η η)Customer Churn Management Campaign AbstractChurn prediction is an important application of classification models that identify those customers most likely to attrite based on their respective characteristics described by e.g. socio-demographic and behavioral variables.Since nowadays more and more of such features are captured and stored in the respective computational systems, an appropriate handling of the resulting information overload becomes a highly relevant issue when it comes to build customer retention systems based on churn prediction models. As a consequence, feature selection is an important step of the classifier construction process. Most feature selection techniques; however, are based on statistically inspired validation criteria, which not necessarily lead to models that optimize goals specified by the respective organization. In this paper we propose a profit-driven approach for classifier construction and simultaneous variable selection based on Support Vector Machines. Experimental results show that our models outperform conventional techniques for feature selection achieving superior performance with respect to business-related goals.

show abstract

Predicting subscriber dissatisfaction and improving retention in the wireless telecommunications industry

Cited by 287 publications

References 4 publications

A diffusion model for churn prediction based on sociometric theory

A diffusion model for churn prediction based on sociometric theory

Impact of Social Network to Churn in Mobile Network

Profit-based feature selection using support vector machines – General framework and an application for customer retention

Contact Info

Product

Resources

About