A Multi-Layer Neural Network Approach to Predict The Success of Bank Telemarketing

Year 2021, Volume: 1 Issue: 1, 69 - 75, 30.04.2021

Mücella Özbay Karakuş

Abstract

In this study, an artificial neural network was constructed and trained with the dataset that created from anonymous data obtained from 45.211 people within the scope of a bank's marketing campaign. The dataset was obtained from an international database, UCI-Irvine Machine Learning Repository. The data set consists of 16 features and a result, the validity of which is based on expert opinion. For the marketing campaigns of the bank, the status of opening a time deposit account at the bank was examined based on the customers’ personal characteristics. For this purpose, the results obtained by using the statistical methods used in the literature for the same data set and the multi-layered artificial neural network (MLNN) were compared. As a result, since the number of data is quite high, it is estimated with a 94.3% higher accuracy whether to open a time deposit account in the bank compared to statistical methods and other artificial neural network methods.

Keywords

artificial intelligence, multi-layer neural marketing, bank telemarketing

References

• 1. Moro S., Cortez P., Rita P. A data-driven approach to predict the success of bank telemarketing. Decision Support Systems; 62, 22-31, 2014.
• 2. Tolon M. Ticari Bankalarda Pazarlama Stratejilerinin Uygulanması ve Türkiye’deki Ticari Bankalar Üzerine Bir Araştırma. Verimlilik Dergisi; 4, 63-88, 2004.
• 3. Baybars B. Türkiye’de Banka Reklamcılığı. İstanbul Üniversitesi İletişim Fakültesi Dergisi; 177-180, 1996.
• 4. Moro S., Laureano R. and Cortez P. Using Data Mining for Bank Direct Marketing: An Application of the CRISP-DM Methodology. Proceedings of the European Simulation and Modelling Conference - ESM'2011; 117-121, 2011.
• 5. UCI-Irvine Machine Learning Repository, International Database, http://archive.ics.uci.edu/ml/datasets/Bank+Marketing (Accessed: April 2021).
• 6. Bahrammirzaee A. A comparative survey of artificial intelligence applications in finance: artificial neural networks, expert system and hybrid intelligent systems. Neural Comput & Applications; 19 (8), 1165-1195, 2010.
• 7. Tomak L., Bek Y. İşlem karakteristik eğrisi analizi ve eğri altında kalan alanların karşılaştırılması. Journal of Experimental and Clinical Medicine; 27 (2), 58-61, 2010.
• 8. Dirican A. Tanı testi performansının değerlendirilmesi ve kıyaslanması. Cerrahpaşa Tıp Dergisi 32, 25-30, 2001.
• 9. Weston J., Watkins C. Support vector machines for multiclass pattern recognition. ESANN’1999, Bruges, Belgium, 219-225, 1999.
• 10. Civalek Ö. Dairesel Plakların Nöro-Fuzzy Tekniği ile Analizi. DEÜ Mühendislik Fakültesi Fen ve Mühendislik Dergisi; 1 (2), 13-31, 1999.
• 11. Güngör M., Fırat M. Askı Madde Konsantrasyonu ve Miktarının Yapay Sinir Ağları ile Belirlenmesi. İMO Teknik Dergi; 219, 3267-3282, 2004.
• 12. Rodrigo M.S.O., Ramon C.F.A., Fabrício J.B.B., Segundo A.P., Zampolo R.F., Fonseca W., Dmitriev V. A System Based on Artificial Neural Networks for Automatic Classification of Hydro-generator Stator Windings Partial Discharges. Journal of Microwaves, Optoelectronics and Electromagnetic Applications; 16 (3), 628-645, 2017.
• 13. Montague G.A., Willis M.J., Tham M.T., Morris A.J. Artificial Neural Network Based Multivariable Predictive Control. Second International Conference on Artificial Neural Networks; 119-123, Newcastle, 1991.
• 14. Hernández G., Hernández L.G., Zamora E., Sossa H., Antelis J.M., Mendoza-Montoya O., Falcón L.E. Using Morphological-Linear Neural Network for Upper Limb Movement Intention Recognition from EEG Signals. Mexican Conference on Pattern Recognition; 389-397, 2019.
• 15. Bakırcı U., Yıldırım T. İleri Beslemeli Ağlarla SPECT Görüntülerinden Kardiyak Problemlerin Teşhisi. IEEE 12. Sinyal İşleme ve İletişim Uygulamaları Kurultayı; 103-105, 2004.
• 16. Özel Y. Farklı Kömür Yakıtları ile Elektrik Enerjisi Üretimi Ve Yakıt Performansının Yapay Zeka Yöntemi Kullanılarak Saptanması. PhD Thesis, Marmara University, Electrical Education, İstanbul, Turkey, 2006.
• 17. Kang H.T., Yoon C.J. Neural Network Approaches to Aid Simple Truss Design Problems. Microcomputers in Civil Engineering; 9, 211-218, 1994.
• 18. Bolat B., Küçük Ü., Yıldırım T. Aktif Öğrenen PNN ile Konuşma/Müzik Sınıflandırma. Akıllı Sistemlerde Yenilikler ve Uygulamaları Sempozyumu; 187-189, 2004.
• 19. Bolat B. Recognition of The Reed Instrument Sounds By Using Statistical Neural Networks. Journal of Engineering and Natural Sciences; 2, 36-42, 2005.