A two-step customer complaint management using ensemble learning for the banking industry

Öz

In this study; It is aimed to classify all kinds of notifications from customers in the banking sector, then prioritize these classes and give feedback to the customer according to this priority. In this way, it is aimed to produce quick solutions for priority notifications that will ensure customer satisfaction. In the literature review on the classification of data, high accuracy values; It has been observed that it is obtained by Logistic Regression, Long Short Term Memory (LSTM), Multinominal Naive Bayes and Support Vector Machine (SVM) algorithms. For this reason, training and testing processes were carried out using Natural Language Processing (NLP) methods on a real bank data set with these algorithms. With the two-stage approach presented as a new method, it has been achieved to increase the accuracy values above seventy percent by working with a limited number of data sets.

Anahtar Kelimeler

• Natural Language Processing
• Machine Learning
• Turkish Text Classification
• Customer Satisfaction
• Logistic Regression
• Support Vector Machine
• Long Short Term Memory
• Multinominal Naive Bayes

Bankacılık sektörü için topluluk öğrenimini kullanan iki aşamalı bir müşteri şikayet yönetimi

Öz

Bu çalışmada; bankacılık sektöründe müşterilerden gelen her türlü bildirimlerin sınıflandırılması, sonrasında bu sınıfların önceliklendirilmesi ve bu önceliğe göre müşteriye geri bildirim verilmesi amaçlanmıştır. Bu sayede müşteri memnuniyeti sağlayacak öncelikli bildirimlere hızlı çözüm üretilebilmesi hedeflenmiştir. Verilerin sınıflandırılmasıyla ilgili yapılan literatür taramasında yüksek doğruluk değerlerinin; Lojistik Regresyon, Uzun Kısa Süreli Bellek, Multinominal Naive Bayes ve Destek Vektör Makinesi algoritmaları ile elde edildiği gözlemlenmiştir. Bu sebeple bu algoritmalarla gerçek bir banka veri seti üzerinde Doğal Dil İşleme (Natural Language Processing - NLP) yöntemleri kullanılarak eğitim ve sınama işlemleri gerçekleştirilmiştir. Yeni bir yöntem olarak sunulan iki aşamalı yaklaşımla sınırlı sayıda veri setiyle çalışılarak doğruluk değerlerini yüzde yetmişin üzerine çıkarılması başarılmıştır.

Anahtar Kelimeler

• Doğal Dil İşleme
• Makine Öğrenimi
• Türkçe Metin Sınıflandırma
• Müşteri Memnuniyeti
• Logistic Regresyon
• Destek Vektör Makinesi
• Uzun Kısa Süreli Hafıza
• Multinominal Naive Bayes

Kaynakça

1. W. Hakiri, “For an efficient complaints management system for banks: A conceptual framework and an exploratory study,” Journal of Marketing Research & Case Studies, vol. 2012, p. 1, 2012.
2. A. Salim, M. Setiawan, R. Rofiaty, F. Rohman et al., “Focusing on complaints handling for customer satisfaction and loyalty: The case of indonesian public banking,” European Research Studies Journal, vol. 21, no. 3, pp. 404–416, 2018
3. X. Lu, Y. Yang, and H. Qin, “The research on personal internet banking service quality and customer loyalty based on complaints handling,” in 2010 International Conference on Management and Service Science. IEEE, 2010, pp. 1–4
4. Z. Eser, M. Pınar, and T. Girard, “Importance of customer complaints: A study of banking industry utilizing the services marketing and branding triangle framework,” Journal of Theory and Practice in Marketing, vol. 2, no. 2, pp. 23–49, 2016
5. S. Harkiranpal, “The importance of customer satisfaction in relation to customer loyalty and retention,” Asia Pacific University, 2006
6. A. Ibrahim, A. Pratiwi, D. I. Meytri, M. A. Kurniawan, N. Yuniarti et al., “Measuring customer satisfaction using crm scorecard in canteen fasilkomunsri,” in 2018 International Conference on Electrical Engineering and Computer Science (ICECOS). IEEE, 2018, pp. 403–408
7. S. Gong, Y. Dai, J. Ji, J. Wang, and H. Sun, “Emotion analysis of telephone complaints from customer based on affective computing,” Computational intelligence and neuroscience, vol. 2015, 2015
8. R. S. A. Corpuz, “Categorizing natural language-based customer satisfaction: an implementation method using support vector machine and long short-term memory neural network,” International Journal of Integrated Engineering, vol. 13, no. 4, pp. 77–91, 2021

9. ——, “An application method of long short-term memory neural network in classifying english and tagalog-based customer complaints, feedbacks, and commendations,” International Journal on Information Technologies and Security, vol. 13, no. 1, p. 2021, 2021
10. R. A. Laksono, K. R. Sungkono, R. Sarno, and C. S. Wahyuni, “Sentiment analysis of restaurant customer reviews on tripadvisor using na ̈ıve bayes,” in 2019 12th International Conference on Information & Communication Technology and System (ICTS). IEEE, 2019, pp. 49–54
11. Z. S. Harris, “Distributional structure,” Word, vol. 10, no. 2-3, pp. 146–162, 1954
12. W. A. Qader, M. M. Ameen, and B. I. Ahmed, “An overview of bag of words; importance, implementation, applications, and challenges,” in 2019 International Engineering Conference (IEC). IEEE, 2019, pp. 200–204
13. A. Kumar, V. Dhanalakshmi, R. Rekha, K. Soman, S. Rajendran et al., “Morphological analyzer for agglutinative languages using machine learning approaches,” in 2009 International Conference on Advances in Recent Technologies in Communication and Computing. IEEE, 2009, pp. 433–435.
14. M. Kintz, C. Dukino, M. Blohm, and M. Hanussek, “Make your customers happy again: Ai and nlp for a customer complaint management platform,” 2020
15. E. Yıldırım, F. S. Çetin, G. Eryiğit, and T. Temel, “The impact of nlp on turkish sentiment analysis,” Türkiye Bilis ̧im Vakfı Bilgisayar Bilimleri ve Mühendisliği Dergisi, vol. 7, no. 1, pp. 43–51, 2015.
16. R. Gavval, V. Ravi, K. R. Harshal, A. Gangwar, and K. Ravi, “Cuda-self-organizing feature map based visual sentiment analysis of bank customer complaints for analytical crm,” arXiv preprint arXiv:1905.09598, 2019.
17. D. Gupta, P. Lenka, H. Bedi, A. Ekbal, and P. Bhattacharyya, “Auto analysis of customer feedback using cnn and gru network,” arXiv preprint arXiv:1710.04600, 2017.
18. D. W. Hosmer Jr, S. Lemeshow, and R. X. Sturdivant, Applied logistic regression. John Wiley & Sons, 2013, vol. 398
19. M. Y. H. Setyawan, R. M. Awangga, and S. R. Efendi, “Comparison of multinomial naive bayes algorithm and logistic regression for intent classification in chatbot,” in 2018 International Conference on Applied Engineering (ICAE). IEEE, 2018, pp. 1–5
20. S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural computation, vol. 9, no. 8, pp. 1735–1780, 1997
21. X. Bai, “Text classification based on lstm and attention,” in 2018 Thirteenth International Conference on Digital Information Management (ICDIM). IEEE, 2018, pp. 29–32.
22. G. Singh, B. Kumar, L. Gaur, and A. Tyagi, “Comparison between multinomial and bernoulli na ̈ıve bayes for text classification,” in 2019 International Conference on Automation, Computational and Technology Management (ICACTM). IEEE, 2019, pp. 593–596
23. C. Cortes and V. Vapnik, “Support-vector networks,” Machine learning, vol. 20, no. 3, pp. 273–297, 1995,
24. Z. Liu, X. Lv, K. Liu, and S. Shi, “Study on svm compared with the other text classification methods,” in 2010 Second international workshop on education technology and computer science, vol. 1. IEEE, 2010, pp. 219–222.