A fraud gang identification and risk propagation prediction method based on a complex network

Abstract

The invention discloses a fraud gang identification and risk propagation prediction method based on a complex network, comprising the following steps: obtaining individual attributes; determining unique encoding of attribute values of the subject attribute and the non-subject attribute; data filtering; establish the data structure of storage and calculation; establishing a connected graph, abstracting the subject attribute value and the non-subject attribute value as nodes, abstracting the belonging relationship between the subject attribute value and the non-subject attribute value as edges connecting the nodes, and composing a connected graph; obtaining a model graph according to the connected graph; calculating model parameters according to the data structure; according to the model parameters, fraud gang identification and fraud risk propagation prediction are carried out. The invention calculates model parameters according to the knowledge of graph theory, and hierarchically classifies the model diagram according to known fraud users, thereby achieving the prediction purpose of accurately identifying fraud groups and fraud risk propagation.

Description

technical field [0001] The invention relates to the field of financial risk control and anti-fraud, in particular to a complex network-based method for identifying fraudulent gangs and predicting risk propagation. Background technique [0002] With the rapid development of Internet finance, the loan process for individuals only requires customers to upload and fill in simple information through the web terminal and pass the approval process before lending. Due to the weak risk control methods of some companies and the default on such financial activities under the current situation The legal accountability of users is still being improved. Some people obtain loan lines through various illegal means, and will not return them after using the line. They become permanently overdue users in the company and bring huge economic losses to the company. These people often It shows the nature of an organized gang, that is, there is a certain relationship between these people, and this ...

AI Technical Summary

Problems solved by technology

[0006] Finally, when calculating the network attributes of gang identification and individual communication, the existing inventions only consider some of the network node attributes, and it is inaccurate to describe the nature of gangs / groups solely from the attributes of the network nodes themselves. Sexual performance is now unable to characterize the overall aggregation of groups

Images