以倒傳遞網路偵測不良信用卡客戶之研究

本篇論文提出一種偵測信用卡不良客戶智慧型推論系統，其中結合如模糊邏輯與倒傳遞網路人工智慧技術來獲取準確的、比傳統風管方法較有效的問題監控者。完善的風險管理方法與有效管理信用風險是銀行長期成功不可或缺要素。信用風險管理在各個機構中與發卡行之間是非常重要地。不良客戶的發生不僅發卡行損失而且影響持卡人在聯合聯信中心的信用狀。然而，大部份實務的壞帳問題是複雜、充滿不確定性，且僅能依人工解決。在這維困的情況下，問題無法有效的預防或克服。在第一階段，我們利用交叉維度來分析輸入值並選取較佳的特徵值。在第二階段，我們試著去找出其他可能的模糊規則特徵值。最後，我們採用所有適合的特微值來實做倒傳遞網路。經由訓練後，利用發卡行提供真實資料來測試系統效能。模擬結果顯示，所提偵測不良信用卡客戶方法可達90%以上精確率。

關鍵字

信用卡、信用風險管理、不良客戶、倒傳遞網路、模糊邏輯

並列摘要

This thesis presents an intelligent inference system to detect bad credit card accounts, where artificial intelligence (AI) techniques such as fuzzy logic (FL) and backpropagation neural network (BPN) will be combined to achieve a more accurate problem detector with a higher availability than those traditional risk management approaches could provide. The effective management of credit risk is a critical component of a comprehensive approach to risk management and essential to the long-term success of any banking organization. Credit-risk-management practices vary considerably among firms and between issuers. The emergence of bad account did not only cause the loss to the issuer but also affect the trustworthiness of cardholders in JCIC. However, most practical problems in bad debts are complex, full of uncertainty and can only be solved by human resources. Due to such a difficult situation, the problems cannot be effectively prevented and overcome. In the first phase, we use cross-table to analyze the value of input nodes and choose better features. In the second phase, we try to find other potential features with fuzzy rule-base. Finally, we adopt all proper features to implement our BPNs. After training, the system performance was tested on real data sets provided by the issuer. Our simulation results reveal that the method for detecting bad credit card accounts reaches accuracy rates of up to 90 percent.

並列關鍵字

backpropagation neural networks ； fuzzy logic ； bad account ； credit-risk-management ； credit card

參考文獻

[6] D. Zhang and L. Zhou, “Discovering golden nuggets: data mining in financial application,” IEEE Trans. on Systems, Man, and Cybernetic—Part C: Applications and Reviews, vol. 34, no. 4, pp.513-522, Nov. 2004.

[8] Y.C. Ye, The Application and Design Pattern of Artificial Neural Networks, Scholars Books Inc., Taipei, Taiwan, 1993.

[9] E. Aleskerov, B. Freisleben and B. Rao, “CARDWATCH: a neural network based database mining system for credit card fraud detection,” in Proc. of IEEE Int. Conf. on Computational Intelligence for Financial Engineering, NY, USA, pp.220-226, March 1997.

[10] V. Chandiramani, R. Jayaseelan, V. Sudhan L. Nathan and K. Priya, “A neural network approach to process assignment in multiprocessor systems based on the execution time,” in Proc. of IEEE Int. Conf. on Intelligent Sensing and Information Processing, Chennai, India, pp.332-335, Aug. 2004.

[11] L.A. Zadeh, “The concept of a linguistic variable and its application to approximate reasoning,” Information Sciences, vol. 8, no. 3, pp.199-249, 1975.

國際替代計量

以倒傳遞網路偵測不良信用卡客戶之研究

未授權

主題瀏覽