Modern Kredi Sınıflandırma Çalışmaları ve Metasezgisel Algoritma Uygulamaları: Sistematik Bir Derleme

Öz Kredi başvurularında, başvuranların temerrüde düşüp düşmeyeceklerinin başarılı şekilde tahmin edilmesi amacıyla önerilen gelişmiş analiz yöntemlerinin sayısı, özellikle Küresel Finans Krizi sonrası dönemde önemli bir artış göstermiştir. Geleneksel istatistiksel sınıflandırma yöntemlerine alternatif olarak bilgiyi, kısıtlar ve varsayımlardan bağımsız olarak doğrudan veri kümelerinden ortaya çıkarma yeteneğine sahip makine öğrenme yöntemleri kullanılmaya başlanmıştır. Bu yöntemlerin yanı sıra, sınıflandırma performansları üzerinde çok büyük iyileştirmeler sağlayan metasezgisel algoritmalar da yazında kendilerine fazlaca yer bulmaya başlamıştır. Veri saklama ve işleme kapasitelerinde yaşanan artıştan en üst düzeyde faydalanmaya yönelik olarak öğrenme yöntemleri ile metasezgisel algoritmaların birlikte kullanımları, kredi risk değerlendirme alanına büyük katkılar sağlamaktadır. Bu derleme kapsamında 2000 sonrası dönemde yazına sunulmuş olan ve metasezgisel algoritmaların yer aldığı kredi sınıflandırma çalışmaları sistematik bir süreç ile incelenmiştir. Yazında karşılaşılan sınıflandırma yöntemleri, uygulanan metasezgisel algoritmalar ile kullanım amaçları ve sınıflandırma performans değerlendirme kriterleri ele alınmış ve mevcut duruma ilişkin genel bir çerçeve oluşturulmuştur. İnceleme, metasezgisel algoritmalar ile makine öğrenme yöntemlerine yönelik artan bir ilgi olduğunu ortaya koymaktadır ancak yöntem tercihleri birkaç alternatif üzerine yoğunlaşmış durumdadır. Yeni geliştirilen metasezgisel algoritmaların ve/veya hibrit ve birlikte kullanımların alanda daha fazla yer alması gerekmektedir. Bilgisayar ve matematik bilimlerinde yaşanan gelişmeler ile paralel olarak ilerleyecek çalışmaların, yazına sürekli katkı sunmaya devam edeceğini söylemek mümkündür.

Metaheuristic Algorithms and Modern Credit Classification Methods: A Systematic Review

Number of proposed advanced analysis methods, which try to successfully predict if applicants are going to default in credit applications show an increasing pattern, especially after the Global Financial Crisis. Alternative to conventional statistical classification methods, machine learning methods arrive on the scene; they have capability to reveal information from the data independently from constraints and assumptions. Along with machine learning methods, metaheuristic algorithms that substantially improves classification performances take part in studies. Combined usages of learning methods and metaheuristic algorithms aim to benefit from the contemporary data storage and process capacities at the highest level and greatly contribute to credit risk assessment field. In this review study, credit classification studies that adopt metaheuristic algorithms in the analyses are examined with a systematic process, for the period after 2000. By forming a general framework, classification methods, metaheuristic algorithm implementations, algorithms’ intended uses and performance assessment criteria are addressed. Examination showed that there is a growing interest, nevertheless method preferences are concentrated over a limited option. It is necessary to incorporate more novel metaheuristics and/or hybrid and combined usages to the studies. It is possible to say that progressive works parallel to the developments in computer and mathematical sciences will continuously contribute to the literature.

___

  • Abdou, H. A. (2009). Genetic programming for credit scoring: The case of Egyptian public sector banks. Expert Systems With Applications, 36(9), 11402–11417. doi:10.1016/j.eswa.2009.01.076
  • Altinbas, H., & Akkaya, G. (2017). Improving the performance of statistical learning methods with a combined meta-heuristic for consumer credit risk assessment. Risk Management, 19(4), 255–280.
  • Boughaci, D. ve Alkhawaldeh, A. A. K. (2018). A new variable selection method applied to credit coring. Algorithmic Finance, 7(1–2), 43–52. doi:10.3233/AF-180227
  • Chen, M. C., & Huang, S. H. (2003). Credit scoring and rejected instances reassigning through evolutionary computation techniques. Expert Systems with Applications, 24(4): 433–441. doi:10.1016/S957-4174(02)00191-4
  • Chen, N., Ribeiro, B., & Chen, A. (2016). Financial credit risk assessment: a recent review. Artificial Intelligence Review, 45(1), 1–23. doi:10.1007/s10462-015-9434-x
  • Chi, B. W., & Hsu, C. C. (2012). A hybrid approach to integrate genetic algorithm into dual scoring model in enhancing the performance of credit scoring model. Expert Systems with Applications, 39(3), 2650–2661. doi:10.1016/j.eswa.2011.08.120
  • Chomboon, K., Chujai, P., Teerarassammee, P., Kerdprasop, K., & Kerdprasop, N. (2015). An empirical study of distance metrics for k-nearest neighbor algorithm. 3rd International Conference on Industrial Application Engineering 2015 içinde (ss. 280–285). doi:10.12792/iciae2015.051
  • Dua, D., & Graff, C. (2017). {UCI} Machine Learning Repository. http://archive.ics.uci.edu/ml adresinden erişildi.
  • Gorzałczany, M. B., & Rudziński, F. (2016). A multi-objective genetic optimization for fast, fuzzy rule-based credit classification with balanced accuracy and interpretability. Applied Soft Computing Journal, 40, 206–220. doi:10.1016/j.asoc.2015.11.037
  • Habibi, A., & Hosseini, S. S. (2016). Ranking bank customers using Neuro-Fuzzy network and optimization algorithms. International Journal of Advanced and Applied Sciences, 3(2), 40–44.
  • Hoffmann, F., Baesens, B., Mues, C., Van Gestel, T. ve Vanthienen, J. (2007). Inferring descriptive and approximate fuzzy rules for credit scoring using evolutionary algorithms. European Journal of Operational Research, 177(1), 540–555. doi:10.1016/j.ejor.2005.09.044
  • Huang, C.-L., Chen, M.-C., & Wang, C.-J. (2007). Credit scoring with a data mining approach based on support vector machines. Expert Systems with Applications, 33(4), 847–856. doi:10.1016/j.eswa.2006.07.007
  • Huang, J. J., Tzeng, G. H., & Ong, C. S. (2006). Two-stage genetic programming (2SGP) for the credit scoring model. Applied Mathematics and Computation, 174(2), 1039–1053. doi:10.1016/j.amc.2005.05.027
  • Huang, S.-C., & Wu, C.-F. (2011). Customer credit quality assessments using data mining methods for banking industries. African Journal of Business Management, 5(11), 4438–4445.
  • Hüllermeier, E. (2005). Fuzzy methods in machine learning and data mining: Status and prospects. Fuzzy Sets and Systems, 156(3), 387–406. doi:10.1016/j.fss.2005.05.036
  • Jadhav, S., He, H., & Jenkins, K. (2018). Information gain directed genetic algorithm wrapper feature selection for credit rating. Applied Soft Computing Journal, 69, 541–553. doi:10.1016/j.asoc.2018.04.033
  • Kaynar, O., Arslan, H., Görmez, Y. ve Işık, Y. E. (2018). Makine öğrenmesi ve öznitelik seçim yöntemleriyle saldırı tespiti. Bilişim Teknolojileri Dergisi, 11(2), 175–185. doi:10.17671/gazibtd.368583
  • Koutanaei, F. N., Sajedi, H., & Khanbabaei, M. (2015). A hybrid data mining model of feature selection algorithms and ensemble learning classifiers for credit scorin. Journal of Retailing and Consumer Services, 27, 11–23. doi:10.1016/j.jretconser.2015.07.003
  • Lanzarini, L. C., Villa Monte, A., Bariviera, A. F., & Jimbo Santana, P. (2017). Simplifying credit scoring rules using LVQ+ PSO. Kybernetes, 46(1), 8–16. doi:10.1108/K-06-2016-0158
  • Liu, X., Fu, H., & Lin, W. (2010). A modified support vector machine model for credit scoring. International Journal of Computational Intelligence Systems, 3(6), 797–804. doi:10.1080/18756891.2010.9727742
  • Louzada, F., Ara, A., & Fernandes, G. B. (2016). Classification methods applied to credit scoring: Systematic review and overall comparison. Surveys in Operations Research and Management Science, 21(2), 117–134. doi:10.1016/j.sorms.2016.10.001
  • Marinaki, M., Marinakis, Y., & Zopounidis, C. (2010). Honey bees mating optimization algorithm for financial classification problems. Applied Soft Computing Journal, 10(3), 806–812. doi:10.1016/j.asoc.2009.09.010
  • Marinakis, Y., Marinaki, M., Doumpos, M., Matsatsinis, N., & Zopounidis, C. (2008). Optimization of nearest neighbor classifiers via metaheuristic algorithms for credit risk assessment. Journal of Global Optimization, 42(2), 279–293. doi:10.1007/s10898-007-9242-1
  • Marqués, A. I., García, V., & Sánchez, J. S. (2013). A literature review on the application of evolutionary computing to credit scoring. Journal of the Operational Research Society, 64(9), 1384–1399. doi:10.1057/jors.2012.145
  • Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & Prisma Group. (2009). Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Annals of Internal Medicine, 151(4), 264–269. doi:10.1371/journal.pmed.1000097
  • Ong, C. S., Huang, J. J., & Tzeng, G. H. (2005). Building credit scoring models using genetic programming. Expert Systems with Applications, 29(1), 41–47. doi:10.1016/j.eswa.2005.01.003
  • Oreski, S., Oreski, D., & Oreski, G. (2012). Hybrid system with genetic algorithm and artificial neural networks and its application to retail credit risk assessment. Expert Systems with Applications, 39(16), 12605–12617. doi:10.1016/j.eswa.2012.05.023
  • Oreski, S., & Oreski, G. (2014). Genetic algorithm-based heuristic for feature selection in credit risk assessment. Expert Systems with Applications, 41(4), 2052–2064. doi:10.1016/j.eswa.2013.09.004
  • Oreški, S., & Oreški, G. (2018). Cost-sensitive learning from imbalanced datasets for retail credit risk assessment. TEM JOURNAL-Technology, Education, Management, Informatics, 7(1), 59–73. doi:10.18421/TEM71-08
  • Qiuju, Z. (2017). Personal credit scoring model research based on the RF-GA-SVM model. Italian Journal of Pure and Applied Mathematics, (38), 235–242.
  • Reddy, K. N., & Ravi, V. (2013). Differential evolution trained kernel principal component WNN and kernel binary quantile regression: Application to banking. Knowledge-Based Systems, 39, 45–56. doi:10.1016/j.knosys.2012.10.003
  • Sun, C., & Jiang, M. (2008). Construction and application of GA-SVM model for prsonal credit scoring. Journal of Information & Computational Science, 5(2), 567–574. doi:10.2495/ameit140271
  • Tsakonas, A., & Dounias, G. (2007). Evolving neural-symbolic systems guided by adaptive training schemes: Applications in finance. Applied Artificial Intelligence, 21(7), 681–706. doi:10.1080/08839510701492603
  • Vukovic, S., Delibasic, B., Uzelac, A., & Suknovic, M. (2012). A case-based reasoning model that uses preference theory functions for credit scoring. Expert Systems With Applications, 39(9), 8389–8395. doi:10.1016/j.eswa.2012.01.181
  • Wang, D., Zhang, Z., Bai, R., & Mao, Y. (2018). A hybrid system with filter approach and multiple population genetic algorithm for feature selection in credit scoring. Journal of Computational and Applied Mathematics, 329, 307–321. doi:10.1016/j.cam.2017.04.036
  • Wang, J., Hedar, A., Wang, S., & Ma, J. (2012). Rough set and scatter search metaheuristic based feature selection for credit scoring. Expert Systems With Applications, 39(6), 6123–6128. doi:10.1016/j.eswa.2011.11.011
  • Zhang, H., He, H., & Zhang, W. (2018). Classifier selection and clustering with fuzzy assignment in ensemble model for credit scoring. Neurocomputing, 316, 210–221. doi:10.1016/j.neucom.2018.07.070
  • Zhou, L., Lai, K. K., & Yu, L. (2009). Credit scoring using support vector machines with direct search for parameters selection. Soft Computing, 13(2), 149–155. doi:10.1007/s00500-008-0305-0