مقایسه عملکرد مدل‌های کلاسیک و هوش مصنوعی در پیش‌بینی وضعیت اعتباری مشتریان بانک

نوع مقاله : مقاله پژوهشی

نویسندگان

1 کارشناسی ارشد گروه مدیریت صنعتی، دانشکده علوم اقتصادی و اداری، دانشگاه مازندران

2 استاد گروه مدیریت صنعتی،دانشکده علوم اقتصادی و اداری، دانشگاه مازندران

چکیده

 در حال حاضر در نظام بانکداری، عدم بازپرداخت تسهیلات به یکی از بزرگ‌ترین مسائل تبدیل شده‌است و به‌دلیل عدم وجود یک سیستم مناسب برای تخصیص تسهیلات، بانک‌ها و موسسات مالی دچار مشکلات عدیده‌ای ازجمله افزایش حجم مطالبات معوق شده‌اند. نظر به اهمیت ریسک اعتباری، بانک‌های تجاری در سطح دنیا درگذشته اغلب از روش قضاوتی برای تعیین ریسک استفاده می‌نمودند، لکن استفاده از این روش‌ها با توجه به توان محدود انسان‌ها در تحلیل هم‌زمان فاکتورهای مختلف مؤثر بر ریسک اعتباری در مقایسه با روش‌های آماری و هم‌چنین روش‌های هوش مصنوعی از کارایی کمتری برخوردار است. به همین منظور این تحقیق درصدد است تا کارایی مدل رگرسیون لجستیک و شبکه عصبی مصنوعی را در تشخیص وضعیت اعتباری مشتریان بانک در فاصله زمانی سال 1388-1392 بسنجد. بررسی نتایج نشان داد که دقت کل مدل شبکه عصبی در داده‌های آموزش 87% و رگرسیون لجستیک 2/77% تعیین شده‌است و خطای نوع اول و دوم در شبکه عصبی به میزان قابل‌ملاحظه‌ای نسبت به روش دیگر کاهش یافته است. با توجه به نتایج نمی‌توان انتظار داشت مدل‌های آماری با مفروضات کلاسیک نظیر خطی بودن روابط متغیرها، بتوانند ریسک اعتباری مشتریان را به درستی ارزیابی نماید؛ از این رو بکارگیری یا تلفیق تکنیک‌های هوش مصنوعی در این مساله ضرورتا توصیه می‌شود.

کلیدواژه‌ها


عنوان مقاله [English]

Comparison of the Performances of Classical Models and Artificial Intelligence in Predicting Bank Customers' Credit Status

نویسندگان [English]

  • Narjes Ghasemnia Arabi 1
  • Abdolhamid Safaei Ghadikolaei 2
1 MSc. Department of Industrial Management, Faculty of Economic and Administrative Science, University of Mazandaran
2 Professor, Department of Industrial Management, Faculty of Economic and Administrative Science, University of Mazandaran
چکیده [English]

Currently, in the banking system, defaults in the repayment of loans have become one of the biggest problems, and banks and financial institutions have faced many problems such as the increase in the volume of outstanding receivables due to the lack of an appropriate system for allocating facilities. Considering the importance of credit risks, commercial banks used to apply judgment methods for determining those risks. However, the use of these methods was not efficient enough due to limited human abilities and, at the same time, various factors affecting credit risks in contrast to statistical methods as well as artificial intelligence methods. For this reason, this article measures the efficiency of logistic regression models and artificial neural networks in detection of bank customers’ credit status in the period of 2009-2013. The results indicated that the total accuracy rates of the Artificial Neural Network model and the Logistic Regression model were 87% and 77.2% respectively, and the error types I and II were reduced significantly in the neural network. According to the results, statistical models cannot be expected to properly evaluate the credit risk of customers with classical assumptions such as the linear relationship between variables. Therefore, application and integration of artificial intelligence techniques is strongly recommended in this regard.

کلیدواژه‌ها [English]

  • Credit scoring
  • Artificial neural network
  • Logistic regression
  • Banking system
1- Ekrami, M. & Rahnama Aski, Azadeh(2009). Investigation the risk factors of past due receivables and deferred Bank. Economic Research Special Bank, 6, 195-216.
2- Alborzi, M., Mohammad Pourzarandi, M., khanbabaei, M. (2010). Using Genetic Algorithm in Optimizing Decision Trees for Credit Scoring of Banks Customers. Journal of Information Technology Management, 2(4), 23-38.
3- Taghavifard, M., Nadali, A. (2012). Classification of Credit Applicants of Banking Systems Using Data Mining and Fuzzy Logic. Industrial Management Studies, 9(25), 85-107.
4- jalili, M., Khodaei valah zadgharad, M. & Koneshlou, M. (2010). real customers Credit scoring in the banking system, Quantitative studies in management,1 (3), 127-148.
5- Dehmordeh, N., Shahraki, J., Seyfoddinpoor, S. & Esfandiyari, M. (2013). Loan Customers Validation Using Credit Scoring model (Case Study : Sepah Bank Branches Zahedan). Management Researches, 5(18), 135-152.
6- Rjabzadeh ghatari, A., Ahmadi, P. & Mirzaei, A. B. (2009).  Designing an Intelligent Combined System for Credit scoring of Bank Customers Using Mixed Fuzzy Argument Models, 53, 195-201.
7- Mirtalaei, M., Azadeh, M. A., Saberi, M. & Ashjari, B. (2010). Hybrid Intelligent Credit Ranking System Using Fuzzy Hybrid-Reasoning Models. Iranian Journal of Trade Studies, 14(53), 159-201.
8- Abdou, H. A., Alam, S. T., & Mulkeen, J. (2014). Would credit scoring work for Islamic finance? A neural network approach. International Journal of Islamic and Middle Eastern Finance and Management, 7(1), 112-125.
9- Abdou, H. A. (2009). An evaluation of alternative scoring models in private banking. The Journal of Risk Finance, 10, 38-53.
 10- Abdou, H. A., & Pointon, J. (2009). Credit scoring and decision m-aking in Egyptian public sector Banks. International Journal of Managerial Finance, 5(4), 391-406.
11- Abdou, H. A., & Pointon, J. (2011). Credit scoring, statistical techniques and evaluation criteria: a review of the literature. Intelligent Systems in Accounting, Finance & Management, 18 (2-3), 59-88.
12- Angelini, E., di Tollo, G., & Roli, A. (2008). A neural network approach for credit risk evaluation. The Quarterly Review of Economics and Finance, 48(4), 733-755.
13- Bekhet, H. A., & Eletter, S. F. K. (2014). Credit risk assessment model for Jordanian commercial banks: Neural scoring approach. Review of Development Finance, 4(1), 20-28.
14- Dimitriu, M., Avramescu, E.A., & Caracota, R. C. (2010). Credit scoring for individuals, Economia. Seria Management, 13(2), 361-377.
15-Dong, G., Lai, K. K., & Yen, J. (2010). Credit scorecard based on logistic regression with random coefficients. Procedia Computer Science, 1(1), 2463-2468.
16- Einav, L., Jenkins, M., & Levin, J. (2013).The impact of credit scoring on consumer lending. The RAND Journal of Economics, 44(2), 27-249.
17- Fogarty, D. J. (2012). Using Genetic Algorithms for Credit Scoring System Maintenance Functions. International Journal of Artificial Intelligence & Applications, 3(6).
18- Harris, T. (2015). Credit scoring using the clustered support vector machine. Expert Systems with Applications, 42(2), 741-750.
19- Hui, L., Li, S., & Zongfang, Z. (2013). The Model and Empirical Research of Application Scoring based on Data Mining Methods. Procedia Computer Science, 17, 911-918.
20- Kozeny, V. (2015). Genetic algorithms for credit scoring: Alternative fitness function performance comparison. Expert Systems with Applications, 42(6), 2998-3004.
21- Lee, T.S., & Chen ,I. F. (2005). A two-stage hybrid credit scoring model using artificial neural networks and multivariate adaptive regression splines. Expert Systems with Applications, 28, 743–752.
22- Li, X. L., & Zhong, Y. (2012). An overview of personal credit scoring: techniques and future work, International Journal of Intelligence Science, 2, 181-189.
23-Liu, Y. (2001). New issue in credit scoring application. Nr Arbeitsbericht, 16, 1-35.
24- Lu, H., Liyan, H., & Hongwei, Z. (2013). Credit scoring model hybridizing artificial intelligence with logistic regression. Journal of Networks, 8(1), 253-261.
25- Marques, A. I., García, V., & Sanchez, J. S. (2012). A literature review on the application of evolutionary computing to credit scoring. Journal of the Operational Research Society, 64(9), 1384-1399.
26- Nguyen, H. T. (2014). Credit Scoring–An empirical study involving auto loans within GE Money Bank France, 1-16.
27- Nicula, I. (2013). Some Aspects Concerning the Measurement of Credit Risk. Procedia Economics and Finance, 6, 668-674.
28- Nurlybayeva, K., & Balakayeva, G. (2013). Algorithmic Scoring Models.Applied Mathematical Sciences, 7(12), 571-586.
29- Sarlija, N., Bensic, M., & Zekic-Susac,M. (2006). Modeling customer revolving credit scoring using logistic regression, survival analysis and neural networks. The 7th WSEAS International Conference on Neural Networks, Cavtat, Croatia, 12-14 June, 164-169.
30- Šušteršicˇ, M., Zupan, J.,&  Mramor, D. (2009). Consumer credit scoring models with limited data. Expert Systems with Applications, 36, 4736–4744.
31-West. D. (2000). Neural Network Credit Scoring Models. Computers and Operations Research, 27, 1131–1152.
32- Yu, L., Wang, S., Lai, K. K., & Zhou, L. (2008). Bio-inspired credit risk analysis: computational intelligence with support vector machines. Springer
33- Zhao, Z., Xu, S., Kang, B. H., Kabir, M. M. J., Liu, Y., & Wasinger, R. (2015). Investigation and improvement of multi-layer perception neural networks for credit scoring. Expert Systems with Applications, 42(7), 3508-3516.