Research On Machine Learning Method For Discriminating Fraud Customers By Credit Investigation Data

As the degree of economic marketization continues to deepen,the proportion of credit fraud in the banking industry is also rising.The prevalence of financially induced fraud has prompted many commercial organizations to put credit information systems in place.The continuous improvement of the credit syste m will help provide more comprehensive credit data to help commercial banks and other financial institutions reduce credit risks and stabilize financial stability.Environment,expand the scope of credit,promote consumer spending,optimize the structure o f economic growth,and promote the sustainable development of financial institutions.In this thesis,data mining technology and machine learning algorithms are used as data analysis methods.Through the comprehensive analysis of individual credit information data in commercial banks and combining algorithms to identify users for fraud,commercial banks can help reduce economic losses.Based on mining the credit data of commercial banks,the paper uses the algorithms in machine learning to model the data through the mining of credit data of commercial banks.This model improves the accuracy,stability and effectiveness of credit card fraud,and reduces the modern credit risk positive effects.