Hybrid Behavioural Features for Churn Prediction in Mobile Telecomm Networks


Behavioural models help to obtain a diagnostic view of the underlying structure to predict a specific action. The importance of the role-played by social ties to better understand the underlying behaviour of customers is recognised. In this work, novel feature of the social aspects of customers’ social group along with the traditional individual customer profiles to extract hybrid behavioural novel features with potential practical implications is incorporated.

As precision plays an important role in churn prediction, one of the main ways of improving prediction performance is with the use of appropriate feature sets that have high predictive power. The prediction model built is based on features that can be extracted from Call Detail Records (CDRs) exclusively. Throughout this work no demographic or contractual information was used, regardless of the data available being restricted to only a period of one month.



> Please use this link to get a copy of the paper

> Please use this link to access the Python codes

> Please use this link to get a copy of the poster


> P. Datta, B. Masand, D. R. Mani, and B. Li, “Automated cellular modeling and prediction on a large scale”, Artif. Intell. Rev., vol.14,no.6, pp. 485–502, December 2000.

> Coussement and D. V. Poel, “Churn prediction in subscription services: An application of support vector machines while comparing two parameter-selection techniques", Expert Syst. Appl., vol. 34, no. 1, pp.313–327, January 2008.

> D. A. Kumar and V. Ravi, “Predicting credit card customer churn in banks using data mining”, Int. J. Data Anal. Tech. Strategies , vol. 1,no. 1, pp. 4–28, August 2008.

> P. Spanoudes and T. Nguyen “Deep Learning in Customer Churn Prediction: Unsupervised Feature Learning on Abstract Company Independent Feature Vectors”, CoRR, abs/1703.03869, March 2017

> A.A.Nanavati,S.Gurumurthy,G.Das,D.Chakraborty, K. Dasgupta, S. Mukherjea, and A. Joshi. "On the structural properties of massive telecom call graphs: findings and implications". In CIKM ’06: Proceedings of the 15th ACM international conference on Information and knowledge management, pages 435–444, New York, NY, USA, ACM, 2006.

> K.Dasgupta,R.Singh,B.Viswanathan,D.Chakraborty, S. Mukherjea, A. A. Nanavati, and A. Joshi. "Social ties and their relevance to churn in mobile telecom networks". In EDBT ’08: Proceedings of the 11th international conference on Extending database technology, pages 668– 677, New York, NY, USA, ACM, 2008

> Y. Richter, E. Yom-Tov, and N. Slonim, “Predicting customer churn in mobile networks through analysis of social groups”, in Proc. SDM 10, Soc. Ind. Appl. Math., Philadelphia, PA, pp. 732–741, 2010.

> V. D. Blondel, J.L. Guillaume, R. Lambiotte, and E. Lefebvre. "Fast unfolding of communities in large networks". In Journal of Statistical Mechanics: Theory and Experiment, page 10008, 2008.