Abstract
Due to the extensive development in the field of telecommunications, so today, it requires that companies must be familiar with and understand the nature of customers and their aspirations. This has led to strong competition between these companies so that they need to use the programmers to build accurate analysis systems to maintain their customers and improve the level of revenue. Here we have introduced an integration system that helps the telecom company achieve that goal. The proposed system consists of three basic pashas: First Phase: An understanding of the company’s data, which consisted of two main parts, included data on the same company in terms of number of employees, number of customers, revenues and expenses, and customer-related data. This phase focuses on initial processing of data that is fragmented and unbalanced. Where the data of the company and the customer was merged first using the joiner and then we addressed the problem of imbalance by building DSMOTE algorithm, which adopted the principle (samples and quadratic) and succeeded in the production of real samples instead of default in the treatment of imbalance. Second Phase: Data were separated after processing into training and testing data. The training data were used to construct a GBM-based predictor after it was developed and replace its decision-making part, which is (DT) with a (GA) algorithm to identify customers to three groups are (the group of customers most influencing the company’s revenues, the group of medium-sized customers and the least important group of customers). Third Stage: The accuracy of the predictor results was verified by using the matrix of the conflict matrix which are: Accuracy, Precision, Recall, F_measure, Fb. A comparison was made between the traditional method of initial treatment, which is SMOTE, DSMOTE in terms of error rate and accuracy. The best results for the developed method were when the data was divided by (40:60) and the error rate was (0.038) and the correct rate (0.962) while the traditional method was the best results for error rate (0.198) and resolution rate (0.802). In addition, the results of the GBM and GBM-GA were compared in terms of the four contrast matrix scales. The traditional method of GBM had the value of Accuracy (0.88), while the developed GBM-GA method was Accuracy (0.97). This confirmed the accuracy of the proposed method.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Wang, H.F. (ed.): Intelligent Data Analysis: Developing New Methodologies Through Pattern Discovery and Recovery. IGI Global, Hershey (2009)
Al_Janabi, S.: Smart system to create an optimal higher education environment using IDA and IOTs. Int. J. Comput. Appl. (2018). https://doi.org/10.1080/1206212X.2018.1512460
Vijaya, J., Sivasankar, E.: An efficient system for customer churn prediction through particle swarm optimization based feature selection model with simulated annealing. Cluster Comput. 1–12 (2017). https://doi.org/10.1007/s10586-017-1172-1
Al_Janabi, S.: A novel agent-DKGBM predictor for business intelligence and analytics toward enterprise data discovery abstract. J. Babylon Univ./Pure Appl. Sci. 23(2), 482–507 (2015)
Jiawei, H., Pei, J., Micheline, K.: Data Mining: Concepts and Techniques, 3rd edn. Elsevier (2011). ISBN 978-0-12-381479-1
Zhu, B., Baesens, B., van den Broucke, S.K.L.M.: An empirical comparison of techniques for the class imbalance problem in churn prediction. Inf. Sci. 408, 84–99 (2017)
Machado, N.L.R., Ruiz, D.D.A.: Customer: a novel customer churn prediction method based on mobile application usage. In: 13th IEEE International Wireless Communications & Mobile Computing Conference, IWCMC 2017, pp. 2146–2151 (2017)
Subramanya, K.B., Somani, A.: Enhanced feature mining and classifier models to predict customer churn for an E-retailer. In: Proceedings of the 7th International Conference Confluence. 2017 Cloud Computing, Data Science & Engineering, pp. 531–536 (2017)
Abd-allah, M.N.: DyadChurn: customer churn prediction using strong social ties, pp. 1–11 (2017)
Chamberlain, B.P., Liu, C.H.B., Pagliari, R., Deisenroth, M.P.: Customer lifetime value prediction using embeddings, pp. 1753–1762. Elsevier (2017)
Milosevic, M., Zivi, N., Andjelkovi, I.: Early churn prediction with personalized targeting in mobile social games. Expert Syst. Appl. 83, 326–332 (2017)
Óskarsdóttir, M., Bravo, C., Verbeke, W., Sarraute, C., Baesens, B., Vanthienen, J.: Social network analytics for churn prediction in telco: model building, evaluation and network architecture. Expert Syst. Appl. 85, 204–220 (2017)
Zhao, L., Gao, Q., Dong, X., Dong, A., Dong, X.: K- local maximum margin feature extraction algorithm for churn prediction in telecom. Cluster Comput. 20(2), 1401–1409 (2017)
Idris, A., Iftikhar, A., Rehman, Z.U.: Intelligent churn prediction for telecom using GP-AdaBoost learning and PSO undersampling. Cluster Comput. 1–15 (2017). https://doi.org/10.1007/s10586-017-1154-3
Vijaya Saradhi, V., Palshikar, G.: Employee churn prediction. Expert Syst. Appl. 38(3), 1999–2006 (2011)
Ali, S.H.: Miner for OACCR: case of medical data analysis in knowledge discovery. In: 2012 6th International Conference on Sciences of Electronics, Technologies of Information and Telecommunications (SETIT), IEEE, Sousse, pp. 962–975 (2012). https://doi.org/10.1109/SETIT.2012.6482043
Al-Janabi, S.: Pragmatic miner to risk analysis for intrusion detection (PMRA-ID). In: Mohamed, A., Berry, M., Yap, B. (eds.) Soft Computing in Data Science. SCDS 2017. CCIS, vol. 788, pp. 263–277. Springer, Singapore (2017). https://doi.org/10.1007/978-981-10-7242-0_23
AlOmari, D., Hassan, M.: Predicting telecommunication customer churn using data mining techniques. In: International Conference on Internet and Distributed Computing Systems, pp. 167–178. Springer, Cham (2016)
Coussement, K., Van den Poel, D.: Improving customer attrition prediction by integrating emotions from client/company interaction emails and evaluating multiple classifiers. Expert Syst. Appl. 36, 6127–6134 (2013)
Owczarczuk, M.: Churn models for prepaid customers in the cellular telecommunication industry using large data marts. Expert Syst. Appl. 37(6), 4710–4712 (2010)
Mansiaux, Y., Carrat, F.: Detection of independent associations in a large epidemiologic dataset: a comparison of random forests, boosted regression trees, conventional and penalized logistic regression for identifying independent factors associated with H1N1pdm influenza infections. BMC Med. Res. Methodol. 14(1), 99 (2014)
Trevor, H., Robert, T., Jerome, F.: The Elements of Statistical Learning, 2nd edn., pp. 337–384. Springer, New York (2009). ISBN 0-387-84857-6
Elith, J., Leathwick, J.R., Hastie, T.: A working guide to boosted regression trees. J. Anim. Ecol. 77, 802–813 (2008)
Robert, N., Gary, M., John, E.: Handbook of Statistical Analysis and Data Mining Applications. Academic Press (2009). ISBN-13: 978-0123747655
Al-Janabi, S., Salman, M.A., Fanfakh, A.: Recommendation system to improve time management for people in education environments. J. Eng. Appl. Sci. 13, 10182–10193 (2018). https://doi.org/10.3923/jeasci.2018.10182.10193
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Al_Janabi, S., Razaq, F. (2019). Intelligent Big Data Analysis to Design Smart Predictor for Customer Churn in Telecommunication Industry. In: Farhaoui, Y., Moussaid, L. (eds) Big Data and Smart Digital Environment. ICBDSDE 2018. Studies in Big Data, vol 53. Springer, Cham. https://doi.org/10.1007/978-3-030-12048-1_26
Download citation
DOI: https://doi.org/10.1007/978-3-030-12048-1_26
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-12047-4
Online ISBN: 978-3-030-12048-1
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)