A method and device for generating anti-fraud features

Abstract

The invention provides a feature extraction method and device that is applied to the construction of anti -fraud models.This feature extraction method first receives the basic information of the applicant and the social information of the applicant, processing the basic characteristics of the applicant's basic information to obtain the basic characteristics, and build social networks based on the applicant's social information, and then use the network to indicate the learning algorithm to calculate the fraud network characteristics of the social network.Finally, the basic characteristics and fraud network characteristics of the stitching applicants are obtained to apply for the characteristics of applicants for anti -fraud modeling.The use of the characteristics of the applicant's applicant to apply for anti -fraud modeling can overcome the inaccurate problem of models caused by only basic information and credit historical information modeling. It can effectively improve the accuracy of the model to identify fraud applicants.Customers apply for credit cards to achieve the pre -prevention of credit card business risks, and help banks establish the first credit risk security network.

Description

technical field [0001] The invention relates to the technical field of machine learning, in particular to a method and device for generating anti-fraud features. Background technique [0002] With the significant economic growth and rapid social progress, my country's credit card industry has developed rapidly. According to the data published by the People's Bank of China in the "Overall Situation of Payment System Operation in 2016", as of the end of 2016, the total number of credit cards and debit cards in use was 465 million; the total amount of bank card credit was 9.14 trillion yuan. In the practice of card issuance, the bank found that along with a large number of normal credit card applications, there are a small number of fraudulent applicants. Such applicants deceive the bank into issuing credit cards by fabricating false personal identity information, falsely using other people's identity information, and providing false certification materials. And withdraw cash,...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the embodiments of the present invention is to provide a method and device for generating anti-fraud features, aiming to solve the problem of low modeling accuracy and poor prediction effect in the prior art using the applicant's basic information and credit history information

Images