Suspicious community discovery method and device, storage medium and computer equipment

Abstract

The embodiment of the invention discloses a suspicious community discovery method and device, a storage medium and computer equipment, which are used for improving the discovery quality and discoveryefficiency of suspicious communities. The embodiment of the invention provides a suspicious community discovery method, which comprises the following steps: constructing a directed weighted network according to transaction snapshots among a plurality of accounts, with each node in the directed weighted network corresponding to one account of the plurality of accounts, and connecting edges of two nodes to indicate that a transaction event occurs between accounts associated with the two nodes, wherein the direction of the edge indicates an execution direction of the transaction event; acquiringa community sequence according to the directed weighted network, wherein the community sequence comprises a plurality of communities extracted from the directed weighted network, and each community comprises a plurality of nodes and edges connecting the two nodes; and obtaining similarity parameters of any two communities from the community sequence, and determining at least two communities of which the similarity parameters exceed a similarity threshold as suspicious communities.

Description

technical field [0001] The embodiments of the present application relate to the field of computer technology, and in particular to a method and device for discovering suspicious communities, a storage medium, and computer equipment. Background technique [0002] In the current social payment scenario, there are transactions between many users, and there are many types of suspicious communities in the transactions between multiple users. For example, typical suspicious communities include: brushing orders, gambling, pyramid schemes, lending and Fraudulent associations of all kinds. In order to ensure the accuracy of strikes, the general strike process includes three steps: discovering suspicious associations, searching for reliable evidence, and attacking suspicious associations. [0003] In the current schemes for discovering suspicious communities, they can be divided into unsupervised suspicious community discovery algorithms and supervised suspicious community discovery ...

AI Technical Summary

Problems solved by technology

For different suspicious types, it is necessary to formulate different suspicious degree calculation methods; for undiscovered risk types, due to lack of relevant business experience, such methods cannot accurately detect suspicious associations

In addition, the unsupervised suspicious community discovery algorithm still has the problem that the quality of the discovered communities is often low. In the social payment scenario, the magnitude of the mined communities is at least one million, so even if 1% of them are manually verified , it also requires a lot of manpower

[0007] For the above-mentioned supervised suspicious community detection algorithm, due to the low coverage of recorded fraudulent accounts as labels, only a small number of suspicious communities can be associated, and there is a problem of low efficiency in finding suspicious communities

Images