Online transaction-oriented fraud detection method based on individual behavior modeling

Abstract

The invention discloses an online transaction-oriented fraud detection method based on individual behavior modeling, and relates to anti-fraud detection of Internet financial network transactions. Themethod is divided into two parts: a first part, generating a heterogeneous information network by using a relational graph and obtaining a vector representation capable of mining a connection betweentransaction attributes by using heterogeneous network representation learning; And in the second part, under the condition that the vector representation of the node is learned, an individual behavior model is established, and the transaction abnormity possibility is predicted. The defects of a traditional fraud detection method are overcome, the mining capacity of the method for potential data connection is improved, and better guarantees are provided for detecting fraud transactions, intercepting the fraud transactions and protecting the fund safety of users and enterprises.

Description

technical field [0001] The invention relates to anti-fraud detection of Internet financial network transactions. Background technique [0002] With the rise of the mobile Internet, various traditional businesses are gradually transferred online. Internet finance and e-commerce are developing rapidly. The generation of online transactions will bring a large amount of electronic transaction data, accompanied by a large number of online payment fraud transactions. Increase. Attackers complete fraud by stealing user accounts, stealing personal privacy information, and even maliciously attacking servers. In order to protect the security of users and the company's business, it is necessary to establish an effective network transaction fraud detection system. [0003] At present, traditional network transaction fraud detection systems usually perform feature transformation for transaction attributes. Fraud detection systems that use these feature transformations often ignore many...

AI Technical Summary

Problems solved by technology

At the same time, there are often intricate potential relationships between transaction data. Only by efficiently capturing the internal potential relationships and maintaining the original structural relationships can more accurate detection of fraudulent transactions occur. This problem has a great impact on the accuracy and robustness of the model. stickiness presents a challenge

In addition, traditional misuse detection mechanisms only try to derive a set of rules for characterizing fraudulent transactions through known fraudulent behaviors, and their main drawback is that they cannot detect novel fraudulent behaviors.

In reality, the tricks of fraudsters continue to evolve, which will make the adaptation of the rules worse and worse

Images