Title :
Echo state network-based credit rating system
Author :
Bozsik, József ; Ilonczai, Zsolt
Author_Institution :
Dept. of Software Technol. & Methodology, Eotvos Lorand Univ., Budapest, Hungary
Abstract :
Nowadays credit rating is of paramount importance in our economic system. In this article a new approach on financial credit rating will be demonstrated. Unlike many classical credit rating methods, the power of this solution lies in the use of artificial intelligence, which is getting more and more prevalent in economics [J. Bozsik, Neural Fuzzy System for Default Forecasts, 11th International Symposium on Computational Intelligence and Informatics, 18-20 Nov. 2010, Budapest, Hungary, pp. 69-74]. The goal has been to create an echo state network-based artificial system, which can provide an effective and reliable way to solve the credit rating task. Besides describing the theoretical background of the echo state network-based solution, this article also focuses on the efficiency of the algorithm and the potential problems in different situations, revealing their solutions as well. As closure, tests will be provided to illustrate the effectiveness of the system.
Keywords :
artificial intelligence; banking; credit transactions; recurrent neural nets; artificial intelligence; echo state network-based artificial system; echo state network-based financial credit rating system; echo state network-based solution; economic system; Accuracy; Biological neural networks; Companies; Neurons; Reservoirs; Testing; Training;
Conference_Titel :
Logistics and Industrial Informatics (LINDI), 2012 4th IEEE International Symposium on
Conference_Location :
Smolenice
Print_ISBN :
978-1-4673-4520-0
Electronic_ISBN :
978-1-4673-4518-7
DOI :
10.1109/LINDI.2012.6319485