Belirsizlik Koşularında Fuzzy Rough Algoritması: Kredi Skorlama’da Bir Uygulama

Yıl 2018, Cilt: 19 Sayı: 2, 55 - 64, 01.07.2018

Nilgün Güler Bayazıt Yasemen Ucan

Öz

Günümüzün artan rekabetçi ortamı, bankaların tüketicilerin artan kredi taleplerine çabuk ve hızlı karar vermelerini gerektirmektedir. Bu amaçla bankalar müşterilere kredi verirken karar vermelerine yardımcı olan istatistik ya da makina öğrenmesi tabanlı kredi skorlama modelleri kullanmaktadırlar. Çalışmada kredi skorlama modellerindeki özellikle belirsizlik konusundaki eksikliği gidermek için bulanık-kaba küme tabanlı bir kredi skorlama modeli önerilmektedir. Bulanık ve kaba kümeler teoremine dayanan yöntem veri kümesindeki örneklerin bulanıklık benzerliklerini hesaplayarak tüketicinin kredi almaya olan uygunluğunu belirleyen kararlar vermektedir. Model sonuçları, yaygın olarak kullanılan diğer kredi skorlama yöntemleriyle karşılaştırılmış ve önerdiğimiz kredi skorlama modellerinden daha iyi olduğunu göstermiştir.

Anahtar Kelimeler

Kredi Skorlama, Bulanık Kümeler, Kaba Kümeler

Fuzzy Rough Set Algorithm under Uncertainty: An Application in Credit Scoring

Öz

The increasing competitive environment in today’s world necessitates a prompt response from the banks to the increasing credit demands of consumers. To serve this purpose, the banks employ statistics or machine learning based credit scoring models that help them in their decision making to give credit to their clients. In this work, a fuzzy rough set based credit scoring model is proposed to remedy the deficiency due to the uncertainty in the credit scoring models. The method is based on fuzzy and rough set theory and makes decisions to determine the suitability of a consumer to receive credit by evaluating the fuzzy similarities of the samples in the data set. The results obtained with the model has been compared with other widely used credit scoring methods and has shown the superiority of our proposed credit scoring method

Anahtar Kelimeler

Creding Scoring, Fuzzy Set, Rough Set

Kaynakça

• Alpaydın, E. (2011). Yapay öğrenme. İstanbul: Boğaziçi Üniversitesi Yayınevi.
• Altman E. (1968). Financial Ratios, Discriminant Analysis and The Prediction of Corporate Bankruptcy. Journal of Finance, 23(4), 589-609.
• Ayanoğlu, Y ve Ertürk, B. (2007). Modern kredi riski yönetiminde derecelendirmenin yeri ve IMKB’ye kayıtlı şirketler üzerinde bir uygulama. Gazi Üniversitesi İktisadi ve İdari Bilimler Fakültesi Dergisi, 9 (2), 75-90.
• Beaver W., (1966), Financial ratios as predictors of failure, Journal of Accounting Research, 4, 71-111.
• Bhatia, B. S. ve Batra, G. S. (1996). Management of Financial Service. New Delhi : Deep & Deep Publications.
• Demirbulut, Y., Aktaş, M., Kalıpsız, O. ve Bayracı S. (2017). İstatistiksel ve makine öğrenimi yöntemleriyle kredi skorlama. Paper presented at UYMS’17. Erişim adresi http://ceur-ws.org/Vol-1980/UYMS17_paper_83.pdf
• Desai, Vijay S., Jonathan N. Crook, and George A. Overstreet. "A comparison of neural networks and linear scoring models in the credit union environment." European Journal of Operational Research 95.1 (1996): 24-37.
• Dubois, D. ve Prade, H. (1990). Rough fuzzy sets and fuzzy rough sets, Internat. J. General Systems, 17 (2–3), 191-209.
• Dua, D. ve Karra Taniskidou, E. (2017). UCI Machine Learning Repository [http://archive.ics.uci.edu/ml]. Irvine, CA: University of California, School of Information and Computer Science.
• Fawcett, T. (2006), An introduction to ROC analysis, Pattern Recognition Letters, 27, 861– 874.
• Hall, M., Frank, E., Holmes G., Pfahringer, B., Reutemann, P., Witten I.H. (2009). The WEKA data mining software: an update. SIGKDD Explorations, (11), Issue1.
• Jensen, R. ve Cornelis, C. (2011). Fuzzy-Rough nearest neighbour classification and prediction. Theoretical Computer Science, 412(42), 5871-5884.
• Kavcıoğlu, Ş. (2014). Ticari bankacılıkta kredi riskinin ve kredi riski ölçüm modellerinin değerlendirilmesi. Finansal Araştırmalar ve Çalışmalar Dergisi, 3 (5), 11-19.
• Ling C.X., Huang J. ve Zhang H. (2003) AUC: A Better Measure than Accuracy in Comparing Learning Algorithms. In: Xiang Y., Chaib-draa B. (eds) Advances in Artificial Intelligence. AI 2003. Lecture Notes in Computer Science (Lecture Notes in Artificial Intelligence), vol 2671. Springer, Berlin, Heidelberg.
• Mester, L. (1997). What's the point of credit scoring?, Federal Reserve Bank of Philadelphia, Business Review, September, 3-16. Erişim adresi https://fraser.stlouisfed.org/files/docs/historical/frbphi/businessreview/frbphil_re v_199709.pdf
• Önder, C. (2010). Bankruptcy prediction with support vector machines. (Yüksek lisans tezi, Humboldt-Universität zu Berlin). Erişim adresi http://dx.doi.org/10.18452/14130
• Pavlak Z. (1982). Rough set. Internat.J.Comput.Inform.Sci., 11(5), 341-356.
• Tabagari, Salome. Credit scoring by logistic regression. Diss. Tartu Ülikool, 2015.
• Radzikowska, A. M., Kerre, E. E. (2002). Comparative study of fuzzy rough sets. Fuzzy Sets and Systems, 125, 137-155.
• Sousa, M. M. ve Reginaldo S. F. (2014). Credit analysis using data mining: application in the case of a credit union. JISTEM-Journal of Information Systems and Technology Management, 11(2), 379-396.
• Walczak, B.,. Massart, D.L. (1999). Rough sets theory. Chemometrics and Intelligent Laboratory Systems, 47, 1-16.
• Yanpeng, Q., Shen, Q., Mac Parthaláin, N., Shang, C. ve Wua W. (2013). Fuzzy similarity-based nearest-neighbour classification as alternatives to their fuzzy- rough parallels. International Journal of Approximate Reasoning, 54, 184-195