| In the past decade,the proportion of telecom network fraud in criminal cases has been increasing year by year.According to the data of the Ministry of Public Security,in some cities with high incidence of fraud cases,telecom fraud cases account for as much as 70%-80%of criminal cases.According to statistics,the number of people receiving fraudulent messages nationwide is as high as 438 million.These figures mean that,on average,1 in every 15 people in our country has received fraudulent phone calls and 1 in every 3 people has received fraudulent text messages.The per capita losses of telecom fraud increased significantly,with the highest amount of fraud per person reaching 210 million yuan.In 2020,the economic loss caused by telecom fraud in China was as high as 22.2 billion yuan.Since the discovery of telecom fraud crimes,the economic loss was no less than 100 billion yuan.In recent years,telecom network fraud crimes continue to occur frequently,the means are increasingly complex,new fraud emerges in an endless stream,and the difficulty of solving the case has been increasing,so systematic research on telecom network fraud for the stable development of banks has important practical significance.This paper adopts literature research,case analysis,and data analysis,on the basis of related literature material finishing,combined with K branch accounts associated with internal and external information of 17891 settlement account account subject from account opening time,sex,age,household registration,opening account in the transaction channel,channel,classifications and net silver IP characteristics is analyzed,To determine the risk factors,and give different risk factors a score according to their impact on telecom fraud,establish a risk early warning model and a dynamic risk monitoring database for hierarchical account management,so as to provide help for commercial banks to identify network fraud.The research result of this paper is:"This paper explores to make up for the defects of existing account risk control means,and tries to study the account’s" early warning "in advance.At the same time,the analysis of real data makes the model and the actual can be highly matched,and the research results can be fed back,which has practical significance;On the other hand,this paper analyzes risk control methods from the perspective that the account is a part of fraud participation,and the research perspective is relatively new.The research results have realized the hierarchical dynamic account control mode of model+score+hierarchy+disposal,so as to achieve timely and effective hierarchical control of accounts.The new account management mode has realized multiple purposes such as actively disposing illegal transactions,enriching management levels,meeting regulatory requirements and improving customers’ property protection capabilities. |