Suspect Customer Identification Of Telecom Fraud

In October 2020,the inter ministerial joint meeting of the State Council on combating new types of illegal and criminal activities in telecommunications networks deployed to carry out the "card breaking" action nationwide,severely crack down on and rectify the illegal and criminal activities of illegal trading of "two cards",and resolutely curb the high incidence of telecommunications fraud.By identifying and controlling new customers,commercial banks are of great help in blocking Telecom fraud and money laundering.Firstly,this paper summarizes the necessity of anti money laundering and briefly introduces the current development status and trend of Telecom fraud in China,and explains the importance of identifying suspicious customers of Telecom fraud in the anti money laundering work of commercial banks from both theoretical and practical aspects.Secondly,using a combination of theoretical and empirical methods,taking the five customer identity indicators recorded by the customers of bank a when opening a new account as the explanatory variable and whether they are suspected of Telecom fraud as the explanatory variable,taking the customer information of the stock customers of bank a who were reported to the people’s Bank of China from 2019 to 2020 as the research data,and using the research data to establish a logistic regression model,Judge whether the above five identity indicators are the significant characteristics of suspected customers of Telecom fraud,and find out the relationship between the above five identity indicators and customers suspected of Telecom fraud,then establish a customer identification model based on BP neural network model,use the information of the above stock customers to train the model and test the effectiveness of the model.Finally,according to the analysis results,from the perspective of commercial banks,this paper puts forward corresponding countermeasures and suggestions for the access of new customers of commercial banks.This study shows that the five customer identity indicators recorded when opening a new account,such as customer gender,age,occupation,annual income and the number of accounts opened,are significant factors suspected of Telecom fraud.Therefore,when opening a new account,we should pay more attention to male customers with low age,low income,high occupational mobility and small number of accounts in bank a;The above five factors are used to establish a customer identification model to identify whether the customers of bank a are suspected of Telecom fraud,and the accuracy of the model can reach more than70%.Therefore,certain control measures are given to the customers identified by the model who may be engaged in Telecom fraud and money laundering,so as to block the telecom fraud in advance and prevent the risk of money laundering.