A trusted subgraph mining method, device and equipment

Abstract

The embodiment of this specification discloses a trusted subgraph mining method, device and equipment. The method includes: obtaining the subgraph corresponding to the graph data to be processed based on the business behavior and / or business relationship of the target point; traversing the subgraph based on the meta-credible recall strategy to obtain an alternative subgraph; A model that classifies the alternative subgraph, and obtains a characterization vector of the alternative subgraph, an overall representation of the alternative subgraph, and a first score of the alternative subgraph, wherein the The first score is used to classify the alternative subgraph to indicate whether the alternative subgraph is credible; based on the characterization vector of the alternative subgraph, the overall representation of the alternative subgraph , using the second model to obtain the second score of the alternative subgraph, wherein the second score represents the risk probability of the alternative subgraph; based on the first score and the second score value, perform subgraph precipitation on the alternative subgraph to obtain a credible subgraph.

Description

Technical field [0001] This specification relates to the field of computer technology, in particular to a method, device and equipment for mining trusted subgraphs. Background technique [0002] With the development of the Internet, advertising on e-commerce platforms, transactions in business processing, web page information, etc., can all be expressed in the form of graph data. In short, information in various scenarios can be converted into graphs for representation . Sub-graphs based on graph data can be used for the precipitation of credible data. The precipitation of credible data can help the release of low-risk transaction events, reduce the amount of analysis at the identification layer, and is of great significance for the identification of credible relationships of transaction events. For the mining of low-risk transaction events or credible data, seek a robust method for quick release, which can not only improve the experience of ordinary users, but also reserve mor...

AI Technical Summary

Problems solved by technology

[0005] The embodiment of this specification provides a trusted subgraph mining method, device and equipment, which are used to solve the following technical problems: the precipitation of trusted data is often limited to certain fixed relationship pairs, lack of scalability, or the precipitation of trusted data In the process, a lot of information will be lost, or the common information of relationship pairs cannot be described, and the problem of dimension explosion is likely to occur when applying multi-subject relationship pairs represented by subgraphs

Images