Multi-standard collaborative fraud detection method based on credit accumulation

Abstract

The invention relates to a multi-standard collaborative fraud detection method based on credit accumulation, and is used for discovering collaborative fraud organizations and identifying the members thereof. The method comprises the following steps that 1) a single suspicious user node is detected and marked in a social network, and the set of all the suspicious user nodes is acquired by using a credit accumulation model; 2) the suspicious relation of the suspicious users in the set delta is judged, and the collaborative fraud organization models to which all the suspicious users belong in the set delta are classified; and 3) the organizers in the classified collaborative fraud organization models are monitored and identified. Compared with the methods in the prior art, the multi-standard collaborative fraud detection method based on credit accumulation has the advantages of being wide in detection objects, advanced in algorithm and comprehensive in assessment.

Description

technical field [0001] The invention relates to the technical field of social network, in particular to a reputation accumulation-based multi-standard cooperative fraud detection method. Background technique [0002] In recent years, with the rapid development of social networks, more and more people use social networks to share emotions, shopping, and some other life experiences with friends or strangers, such as microblogs and blogs. Reputation is the basis for reflecting past creditworthiness and predicting future personal reliability. It has become an effective and feasible solution for network security management. A reliable reputation system can determine whether members in the community are trustworthy. However, unscrupulous users or merchants seek profits by taking advantage of their own high reputation or maliciously destroying the reputation of honest users. For example, unscrupulous users in social networks use anonymous attacks on other honest users and defraud ...

AI Technical Summary

Problems solved by technology

Current research mainly focuses on the detection of individual user fraud, without considering collaborative fraud among users, and in the real-world application of social networks, collaborative fraud among users is more damaging than individual user fraud and higher frequency

Images