A Survey on Churn Prediction Techniques in Communication Sector

Abstract

The speedy augmentation of the market in every sector is leading to superior subscriber base for service providers. Added competitors, novel and innovative business models and enhanced services are increasing the cost of customer acquisition. In such a tedious set up service providers have realized the importance of retaining the on hand customers. It is therefore mandatory for the service providers to inhibit churn-a phenomenon which states that customer wishes to quit the service of the company. To prevent the churn many approaches are used by the researchers. This paper reviews the different approaches used by researchers not only in communication sector but also other sectors which highly depends on customer participation.

Analytics: Key to Go from Generating Big Data to Deriving Business Value

“…The problem of predicting churn and non-churn customers has been addressed in number of studies [31], [32]. However, with increasing competition, the companies are now turning towards machine learning algorithms to gain early insights about their customers' behavior such that timely actions can be taken to prevent customer churn.…”

“…One simple approach to predict if the user is churn or non-churn customer, is to formulate it as a two class classifier problem using underlying feature values to predict the outcome. Some of the possible features that can be used to define churn and non-churn classes, includes duration of customers calls, services subscribed, usage pattern, and demographics [31]. A comprehensive review of the approaches that can be followed to predict churning customer is presented in [31], [32].…”

“…Since cost of acquiring a new customer is relatively high compared to retaining the existing customers [31], companies are developing new and competitive ways to retain their customers and maintain long term relationship with them to avoid customer churn. Churners are the customers who leave their existing telecommunication service provider and switch to new ones for different reasons [32].…”

“…While big data analyses provide value to businesses there are issues surrounding it in general that must be carefully dealt with to exploit its full potential [31], [1]. One of the primary concerns around big data is security and privacy.…”

Survey of Data Mining Techniques Used for Real Time Churn Prediction

“…P.C. Pendharkar [18] proposed genetic-algorithm based neural network model to predict churn and compared the result with statistical z-score based prediction model. Javad Basiri et al [19] proposed a hybrid approach (OWA) based on LOLIMOT and Bagging & Boosting algorithms to improve the prediction accuracy of churn and used chi-square algorithm for feature selection.…”