Advertisement

An Application in Bank Credit Risk Management System Employing a BP Neural Network Based on sfloat24 Custom Math Library Using a Low Cost FPGA Device

  • Maria Cristina Miglionico
  • Fernando Parillo
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 300)

Abstract

Artificial Neural Networks (ANNS) base their processing capabilities in parallel architectures. This makes them useful to solve pattern recognition, system identification and control problems. In particular, it is extremely important for commercial banks to set up an early bank credit risk warning system. The authors set up early warning indicators for commercial bank credit risk, and carry out the warning for the credit risk in advance with the help of the ANNS.

A three layer ANN has been implemented, using a custom developed sfloat24 math library, on a low cost FPGA device.

Keywords

Artificial Neural Network Field Programmable Gate Array ((FPGA) sfloat24 math library 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Banuelos, M.A., Castillo Hernandez, J., Quintana Thierry, S., Damian Zamacoma, R., Valeriano Assem, J., Cervantes, R.E., Fuentes Conzalez, R., Calva Olmos, G., Perez Silva, J.L.: Implementation of a Neuron Using FPGAS. Journal of Applied Research and Technology I(003), 248–255 (2003)Google Scholar
  2. 2.
    Lapskey, P., Bier, J., ShoHam, A., Lee, E.A.: DSP Processor Fundamentals – Architecture and features. IEEE Press (1997)Google Scholar
  3. 3.
    Zhao, S.-F., Chen, L.-C.: The BP Neural Networks applications in Bank Credit Risk Management System. In: 8th IEEE International Conference on Cognitive Informatics, ICCI 2009 pp. 527–532 (2009)Google Scholar
  4. 4.
    Wang, Y., Yan, H., Meng, X.: Matching Decision Model for Self-adaptability of Knowledge manufacturing System. In: International Conference on Information Science and Technology (ICIST), pp. 891–895 (2011)Google Scholar
  5. 5.
    Zhou, H., Wang, J., Qiu, Y.: Application of the Cross Entropy Method to the Credit Risk Assessment in an Early Warning System. In: International Symposiums on Information Processing (ISIP), May 23-25, pp. 728–732 (2008)Google Scholar
  6. 6.
    Quer, G., Meenakshisundaram, H., Tamma, B., Manoj, B.S., Rao, R., Zorzi, M.: A Cognitive Network Inference through Bayesan Network Analysis. In: 2010 IEEE Global Telecommunications Conference (GLOBECOM 2010), December 06-10, pp. 1–6 (2010)Google Scholar
  7. 7.
    Pang, X.-L., Feng, Y.-Q.: An Improved Economic Early Warning Based on Rough Set and Support Vector Machine. In: International Conference on Machine Learning and Cybernetics, August 13-16, pp. 2444–2449 (2006)Google Scholar
  8. 8.
    Miglionico, M.C., Parillo, F.: Modelling a neuron using a custom math library sfloat24 – Implementation of a sigmoid function on a FPGA device. In: Proceedings of the International Symposium on the Analytic Hierarchy Process for Multicriteria Decision Making, ISHAP Conference, Sorrento Italy, June15-18 (2011) ISSN 1556-8296, http://204.202.238.22/isahp2011/dati/autor.html
  9. 9.
    Miglionico, M.C., Parillo, F.: A Current Hysteresis Controller for Reduction of Switching Losses in a Full-Bridge Inverter – FPGA implementation by using a custom developed 24 bit Floating Point Math Library. In: IEEE Conference UPEC 2011, Soest, Germany, September 05-08, pp. 1–6 (2011)Google Scholar
  10. 10.
    Miglionico, M.C., Parillo, F.: FPGA implementation of sfloat24 digital PI. In: IEEE Conference PEDES 2010 Power India, New Delhi, December 20-23 (2010)Google Scholar
  11. 11.
    IEEE Standard for Binary Floating-Point Arithmetic, ANSI/IEEE 754 (1985)Google Scholar
  12. 12.
    Kahan, W.: Lecture notes on the Status of IEEE Standard 754 for Binary Floating Point Arithmetic. Electrical Engineering and Computer Science. University of California, Berkeley (May 31,1996)Google Scholar
  13. 13.
    Attaianese, C., Parillo, F., Tomasso, G.: Dual Boost High Performances control strategy on a Power Factor Correction (PFC) implementation by using a 24 bit custom floating point library. Journal of Electrical Engineering 10 (December 23, 2010), http://www.jee.ro
  14. 14.
    Ginsberg, S.: Compact and Efficient Generation of Trigonometric Functions using a CORDIC algorithm, Cape Town, South Africa (January 2002)Google Scholar
  15. 15.
    Miglionico, M.C., Parillo, F.: A BP Neural Network Application in Bank Credit Risk Management System using a sfloat24 custom math library – FPGA implementation. In: A.M.A.S.E.S. Meeting, XXXV edn., Pisa, Italy, September 15-17 (2011)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Maria Cristina Miglionico
    • 1
    • 2
  • Fernando Parillo
    • 3
  1. 1.Department of Culture of the ProjectSecond University of NapoliAversaItaly
  2. 2.BENECON ScarlUNESCOItaly
  3. 3.Department of Electrical Engineering and InformationUniversity of CassinoCassinoItaly

Personalised recommendations