A Hybrid Approach to Predict Churn

Abstract

Acquiring new customers in any business is much more expensive than trying to keep the existing ones. As a result, many prediction algorithms have been proposed to detect churning customers. In this paper, the ordered weighted averaging (OWA) technique is brought to the attention of marketing researchers. We have applied OWA technique to improve the prediction accuracy of existing churn management systems. The decision lists of underlying prediction algorithms have been fused using OWA algorithm. Applied to the database of a telecommunication company, this method is found to significantly improve accuracy in predicting churn compared to the best existing result in the literature of the churn management. Our findings lead us to believe that using OWA technique could cause to increase profit for the companies.

Churn Prediction in Telecommunication Industry using Decision Tree

“…Classification is one of the data mining tasks that focuses on classifying unknown cases based on a set on known examples [9]. Hence, classification techniques based on data mining can be used for predicting churn in telecommunication industry [8] [10].Wai-Ho Au et al [3], Erfaneh and Tarokh [4], Wei Yu et al [5], Chao et al [6], Adnan and Asifullah [7] have also focused on using data mining techniques for churn prediction in telecommunication in their work. Various techniques have been used by various researchers but use of data mining techniques for predicting customers churn has turned out to be an efficient approach with high accuracy in results.…”

A Survey on Churn Prediction Techniques in Communication Sector

“…Fig. 1 illustrates the share of each algorithm in the churn management systems [9,21]. Based on reference materials we can come to the conclusion that the decision tree, neural network, regression and cluster analysis are preferred by the most of researchers where as the other algorithms are not used to that significant level.…”

“…P.C.Pendharkar [12] 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, [21] 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. The Order weighted averaging (OWA) method uses the strength of both LOLIMOT and bagging and boosting classification tree.…”

Survey of Data Mining Techniques Used for Real Time Churn Prediction

“…Owczarczuk [12] used logistic regression, G. Nie et al [13] used logistic regression and decision tree model and A. Keramati, S.M.S. Ardabili [14] focused on Binomial logistic regression model for churn prediction and identified customer dissatisfaction, service usage, switching cost and demographic variable affects customer churn. B. Shim et al [15] used decision tree, neural network and logistic regression for customer classification and identified decision tree shows highest hit ratio among them and P. Kisioglu, Y.I.…”