Design And Implementation Of Commercial Bank Risk Control System

Just as finance is the core of modern economy,banks are the main body of modern financial system;information technology is the development trend of modern society.Meanwhile,data is thesupporter of information.As we all know,the global economic integration is accelerating,therefore,higher level management and technique are on request for banking industry.E-business,ERP fordata management and decision will be the future development priority.Banks are a high-income industry with high risk;it is exposed to all kinds of risks all the time.How to control the risk and opportunity are the problems confronting the banks.So risk management cannot be a remedy after the event,but a prerequisite for pre-action behavior,the aim of which is to avoid the occurrence of crisis.Commercial banks now have a large number of customer base,hence,a large amount of data were accumulated in the long-term business.Reasonable application of those data,you could find valuable one and do the risk precaution,in the meanwhile,you improve the maximum utilization of those data.Based on the risk control system developed jointly by the company and a commercial bank case,this paper used the bank's existing CRM data base to extract the risk-related business data and build the risk data mart.On this basis,I made a detailed analysis to credit risk and managed credit risk classification.Using this data mining technology to tap the hidden risks for the bank,I think it could help to predict the possible risks with a more scientific way.The main work is as follows.1)This paper provides a design proposal of the overall structure of the risk control system for commercial banks.By analyzing the business needs of commercial banks,the function modules of the system are divided.The functions are mainly data preprocessing,credit risk multidimensional analysis and risk weighting.According to the system function,the hierarchical structure of the risk control system,the main data source layer,the data preparation layer,the analysis processing layer and the front application layer are proposed.Base on the process of data processing,the data can be divided into risk data mart,data interface,data processing and report display.2)Multidimensional Analysis of Credit Risk Data.Usingthe Teradata Administrator tool and the SQL statement to extract the relevant risk data into the temporary table,I split thetheme and form multiple dimension tables to build the multidimensional analysis model.We take customers,industry,assets as the core to find the potential risks of default of loan customerswith multi-angle analysis.3)Credit Risk Classification.The aim of this part is to research how to construct a decision tree based on CART algorithm,and establish a credit classification model.By analyzing the lender's repayment intention and repayment ability,we could grab high quality customers and potential risk customers for banks.Risk control system is currently running stable.We couldprovide visual query analysis to meet the needs of banks bydeveloping new reports and regularly pushing the data to the report module.