The Application Of Genetic Algorithms In Commercial Bank Credit Risk Assessment

Commercial banks occupies a crucial position in the financial system, and have played a central role in the implementation of monetary policy and social investment. At the same time, commercial banks also faced with a variety of financial risks in the course of its business. These risks will not only affect the operating results of commercial banks, but also determines the survival of commercial banks. As the China's economic reform and accession to WTO economic globalization is gradually deepening, the banking industry in turn is faced with many new and unprecedented risks. How to handle these new risks will have a great significance to maintain the financial system's stability and economic security. In this context, these articles do research into on commercial bank risk assessment by using the genetic algorithms.First we introduced the commercial bank risk. Mainly introduced the origin and development of the commercial bank risk management. And we discussed several types of risks which commercial banks faced and current management methods in the world, and raised the main problems and goals about China's commercial bank risk management.Secondly, the article describes the background of rough set theory and research status. Introduced the basic concepts of Rough set theory in details, explained the basic idea of attribute reduction through examples. Make a briefly described about the original sources date used in experiment, introduced the two kinds of method of data discretization we used. At last we give a description about the training data after discretization processed and a specific algorithm which was used for the data discretization.Finally, we introduced the characteristics of genetic algorithms and the basic principles and processes. The article makes an analysis about the various elements of the genetic algorithm that impact the performance of the genetic algorithm. This paper made studies about how to use the genetic algorithm to solve the commercial bank's risk assessment. Proposed a total solution for discretization of the original data and attribute reduction, obtain the classification rule by calling the genetic algorithm. Carried out an improvement in the data pre-processing and the genetic manipulation, improved the convergence speed and search efficiency of genetic algorithm. The final experimental results show that the whole program is feasible and effective.