Abstract

The need for customer satisfaction and continuity is very important in the banking sector as in other sectors. Increasing number of services and products are leading to increased competition in the sector and this causes customers to be more anticipated towards service providers. With globalization, banks have had to make new decisions for customer retention and keeping up with change. Implementing their new decisions requires strengthening their infrastructures and technologies and passing customer-focused systems. In this study, the characteristics of the customers who can leave firms in the banking sector are tried to be determined by using the information of the previous customers. The algorithm that makes the most accurate prediction was tried to be determined by applying ZeroR, OneR, Naive Bayes, J48 and Multilayer Perceptron algorithms to the bank customer data using Weka, one of the data mining software. According to the results of the algorithm, the reasons why the customers leave the bank were tried to be identified.

References

[1] Bryan F, Stone M, CRM in Financial Services: A Practical Guide to Making CustomerRelationship Management Work, Kogan Page Limited, Milford, Ct, USA, 2002.
[2] Eichorn FL, Internal Customer Relationship Management (IntCRM) A Framework forAchieving Customer Relationship Management from the Inside Out, Problems andPerspectives in Management 1, (2004), ss. 154 177.
[3] Nancarrow C, Rees S, Stone M, New Directions in Customer Research and the Issue ofOwnership: A Marketing Research Viewpoint, Database Marketing- Customer StrategyManagement, (2003), Vol. 11, ss. 26 39.
[4] Koyuncugil, A, S, Özgülbaş N, Veri Madenciliği: Tıp ve Sağlık Hizmetlerinde Kullanımıve Uygulamaları, bilişim teknolojileri dergisi, cilt: 2, sayı: 2, Mayıs 2009
[5] Emel G.G. Taşkın Ç., Veri Madenciliğinde Karar Ağaçları ve Bir Satış AnaliziUygulaması, Eskişehir Osmangazi Üniversitesi Sosyal Bilimler Dergisi Cilt: 6 Sayı: 2Aralık 2005
[6] Savaş S., Topaloğlu N., Veri Madenciliği Yöntemi İle GSM Şebekelerinin PerformansAnalizi, Gazi Üniv. Müh. Mim. Fak. Der. Cilt 26, No 4, 741-751, 2011
[7] Terzi S., Hile ve Usulsüzlüklerin Tespitinde Veri Madenciliğinin Kullanımı, The Journalof Accounting and Finance April / 2012
[8] Akçetin E., Çelik U., The performance benchmark of decision tree algorithms for spame-mail detection , İnternet uygulamaları ve Yönetim Dergisi, Cilt 5, Sayı 2, 2014
[9] Narli S., Aksoy E., Ercire Y E., Investigation of Prospective Elementary MathematicsTeachers’ Learning Styles and Relationships between Them Using Data Mining,International Journal of Educational Studies in Mathematics, 2014, 1 (1), 37-57, ISSN:2148-5984
[10] Kalıkov A, Veri Madenciliği ve Bir E-Ticaret Uygulaması, Yüksek Lisans Tezi, GaziÜniversitesi, Fen Bilimleri Enstitüsü, 2006
[11] İnan, O, Veri Madenciliği, Yüksek Lisans Tezi, Selçuk Üniversitesi, Fen BilimleriEnstitüsü, 2003.
[12] Han J, Kamber M, Pei j, Data Mining Concepts and Techniques, Morgan Kaufmann,Elsevier, London 2001
[13] Delen D, Walker G, Kadam A, Predicting breast cancer survivability: a comparison ofthree data mining methods, Artificial Intelligence in Medicine, 34(02), s (113-127),2005.
[14] Witten I H, Frank E, Hall M A, Data mining: practical machine learning tools andtechniques, ISBN: 9780128043578, Morgan Kaufmann, Elsevier, London, 2011.
[15] Seker S E, Weka İle Veri Madenciliği, ISBN: 9781311400574, Bilgisayar KavramlarıYayınları, Ataşehir, İstanbul, 2015.
[16] Öztemel E, Yapay Sinir Ağları, Papatya Yayınları, s.29, 2006.
[17] Patterson D, Liu F, Turner, D, Concepcion A, Lynch, R, Performance Comparison ofthe Data Reduction System, Proceedings of the SPIE Symposium on Defense andSecurity, Mart, Orlando, FL, 2008.
[18] Hall M, Frank E, Holmes G, Pfahringer B, Reutemann P, Witten I H, The WEKA DataMining Software: An Update, ACM SIGKDD Explorations Newsletter, 11(1), 10–18,200