Bank Financial Business Risk Assistant Management System Based on Clustering Algorithm

Abstract

In recent years, compliance risk has gradually become an important risk faced by China's commercial banks in addition to credit risk, market risk and operational risk. The lifeline of commercial banks is risk management. In view of the harmfulness of the financial crisis, this paper tries to find out the relevant index parameters that have obvious impact on the financial risk. By building a prediction model, it can predict the risk that may have an impact on China, effectively find the potential financial risk, serve the operation of the macro financial market, and improve the accuracy of financial supervision. In this paper, the current web crawler technology and clustering data mining analysis technology are applied to build an efficient audit risk management system based on bank financial risk management. It takes a lot of time, manpower and energy to screen the original messy and high-risk huge basic data, which effectively improves the management and control efficiency of audit risk in the current social security field, it reduces the workload of audit risk identification and control in the audit process, and can be popularized in some areas. Based on the construction of chaotic neural network prediction model based on wavelet clustering algorithm and FCM clustering algorithm, this paper makes an empirical comparison of the early warning effect of the two models, and compares the advantages and disadvantages of adding different clustering algorithms and chaotic neural network prediction through empirical comparison. After model comparison and analysis, it is found that the risk prediction system of chaotic neural network prediction model based on wavelet clustering algorithm has high accuracy, and the maximum error is only 0.1173, which meets the needs of users.