Construction method of social network fraud model, fraud prevention method and device

Abstract

The invention relates to a construction method of a social network fraud model, a fraud prevention method and a fraud prevention device. The method comprises the following steps: inputting social network information of a training object into a pre-constructed social network extraction model to obtain feature data of the social network information so as to construct a social network map corresponding to the social network information; identifying the social network map based on a preset community identification algorithm to obtain the number of communities corresponding to the social network map; obtaining historical behavior data of each community object in each community; determining whether the target community has fraud behaviors or not according to all historical behavior data in the target community; and if the target community has the fraud behavior, training a pre-constructed neural network model by utilizing all historical behavior data in the target community to obtain a social network fraud model so as to identify the fraud behavior of the target network information by utilizing the social network fraud model. The fraud behavior of the target network information can be quickly and accurately identified and predicted.

Description

technical field [0001] The invention relates to the technical field of consumer finance, in particular to a method for constructing a social network fraud model, a fraud prevention method and a device. Background technique [0002] In recent years, social networks have developed rapidly, and more and more people have begun to use them. With the popularity of social networks such as Weibo, WeChat, and Facebook, social networks account for an increasing proportion of users' Internet life. Unfortunately, the easy availability of personal information freely shared by people on social networks makes it relatively easy for fraudsters to steal user credentials and identity information to defraud consumers. Therefore, in order to effectively control the risk of fraud It is necessary to analyze the complex social relationships in the social network to find out whether there is fraudulent behavior in the social network in time. [0003] The traditional method of social network analy...

AI Technical Summary

Problems solved by technology

[0004] However, the existing technology is limited by subjective and objective conditions in the process of social network analysis, and the observation and summary of human experts cannot be completely comprehensive and accurate; and as the actual scene changes, it often leads to a new round of features engineering iteration

For example, when the learning task goal changes, or the original data distribution changes, and the arrival of new data, the old features will no longer be applicable to the new environment, which requires redesign and selection of new features, which takes a long time

Images