Method and apparatus for identifying risky vertices

Patent metadata

Field	Value
Publication number	US-2021357942-A1
Application number	US-202117323549-A
Country	US
Kind code	A1
Filing date	May 18, 2021
Priority date	Mar 8, 2019
Publication date	Nov 18, 2021
Grant date	—

Abstract

The present specification provides a method for identifying risky vertices, including: constructing multiple medium networks, each of the multiple medium networks being constructed from one or more black seeds of the same type and including vertices and media connected to the vertices; determining a first risk value of each vertex based on a quantity of upper-layer media connected to the vertex and a quantity of risk conditions that the vertex meets; determining a final risk value of each vertex based on a quantity of overlapping times of the vertex in a stacked medium network structure and the first risk value; and determining a high-risk vertex based on the final risk value.

First claim

1 . A method for identifying a risky vertex, comprising:
constructing a plurality of medium networks, each of the medium networks being constructed from one or more black seeds of a same type and including vertices and media connected to the vertices; determining a first risk value of each vertex based on a quantity of one or more upper-layer media connected to the vertex and a quantity of one or more risk conditions that the vertex meets; stacking the plurality of medium networks to generate a stacked medium network structure; determining a final risk value of each vertex based on a quantity of one or more overlapping times of the vertex in the stacked medium network structure and the first risk value; and determining a high-risk vertex based on the final risk value.

2 . The method according to claim 1 , wherein the constructing the plurality of medium networks includes:
generating one or more initial seeds; matching media in a medium pool with the one or more initial seeds to generate one or more initial media; and outputting the one or more initial seeds and the one or more initial media to construct the medium networks.

3 . The method according to claim 2 , wherein the matching the media in the medium pool with the one or more initial seeds includes:
determining whether each medium of the media is associated with one or more of the one or more initial seeds; determining a ratio between a quantity of the one or more initial seeds associated with the medium and a total quantity of the one or more initial seeds; and determining the medium to be an initial seed if the ratio is greater than a threshold.

4 . The method according to claim 1 , wherein the determining the first risk value of each vertex based on the quantity of one or more upper-layer media connected to the vertex and the quantity of one or more risk conditions that the vertex meets includes:
determining an initial risk value of each vertex based on the quantity of one or more upper-layer media connected to the vertex; and determining a second risk value of each vertex based on the quantity of one or more risk conditions that the vertex meets and the initial risk value.

5 . The method according to claim 4 , wherein the initial risk value of the vertex is a logarithmic function of the quantity of one or more upper-layer media connected to the vertex.

6 . The method according to claim 4 , wherein the determining the second risk value of each vertex based on the quantity of one or more risk conditions that the vertex meets includes:
determining the quantity of one or more risk conditions that the vertex meets; and determining the second risk value of the vertex based on the quantity of one or more risk conditions that the vertex meets and the initial risk value.

7 . The method according to claim 6 , wherein the determining the second risk value of the vertex based on the quantity of one or more risk conditions that the vertex meets and the initial risk value further includes:
determining a layer number of the vertex; and determining the second risk value of the vertex based on the layer number of the vertex, the quantity of one or more risk conditions that the vertex meets, and the initial risk value.

8 . The method according to claim 7 , wherein the layer number of the vertex is a quantity of media between the vertex and at least one of the one or more black seeds.

9 . The method according to claim 7 , further comprising:
comparing a risk value of each boundary vertex with risk values of upstream vertices of the boundary vertex; and responsive to determining that the risk value of the boundary vertex is greater than a risk value of any of the upstream vertices, adjusting the risk values of the upstream vertices of the boundary vertex to cause each of the risk values of the upstream vertices to be greater than or equal to the risk value of the boundary vertex.

10 . The method according to claim 1 , wherein the obtaining the final risk value of each vertex in the stacked medium network includes:
increasing a risk value of an overlapped vertex based on a quantity of overlapping times of the vertex.

11 . An apparatus for identifying risky vertices, comprising:
a module configured to construct a plurality of medium networks, each of the medium networks being constructed from one or more black seeds of a same type and including vertices and media connected to the vertices; a module configured to determine a first risk value of each vertex based on a quantity of one or more upper-layer media connected to the vertex and a quantity of one or more risk conditions that the vertex meets; a module configured to stack the multiple medium networks to generate a stacked medium network structure; a module configured to determine a final risk value of each vertex based on a quantity of one or more overlapping times of the vertex in the stacked medium network structure and the first risk value; and a module configured to determine a high-risk vertex based on the final risk value.

12 . The apparatus according to claim 11 , wherein the module configured to construct the plurality of medium networks includes:
a module configured to generate one or more initial seeds; a module configured to match media in a medium pool with the one or more initial seeds to generate one or more initial media; and a module configured to output the one or more initial seeds and the one or more initial media to construct the medium networks.

13 . The apparatus according to claim 12 , wherein the module configured to match the media in the medium pool with the one or more initial seeds includes:
a module configured to determine whether each medium of the media is associated with one or more of the one or more initial seeds; a module configured to determine a ratio between a quantity of the one or more initial seeds associated with the medium and a total quantity of the one or more initial seeds; and a module configured to determine the medium to be an initial seed if the ratio is greater than a threshold.

14 . The apparatus according to claim 11 , wherein the module configured to determine the first risk value of each vertex based on the quantity of one or more upper-layer media connected to the vertex and the quantity of one or more risk conditions that the vertex meets includes:
a module configured to determine an initial risk value of each vertex based on the quantity of one or more upper-layer media connected to the vertex; and a module configured to determine a second risk value of each vertex based on the quantity of one or more risk conditions that the vertex meets and the initial risk value.

15 . The apparatus according to claim 14 , wherein the initial risk value of the vertex is a logarithmic function of the quantity of one or more upper-layer media connected to the vertex.

16 . The apparatus according to claim 14 , wherein the module configured to determine the second risk value of each vertex based on the quantity of one or more risk conditions that the vertex meets includes:
a module configured to determine the quantity of one or more risk conditions that the vertex meets; and a module configured to determine the second risk value of the vertex based on the quantity of one or more risk conditions that the vertex meets and the initial risk value.

17 . The apparatus according to claim 16 , wherein the module configured to determine the second risk value of the vertex based on the quantity of one or more risk conditions that the vertex meets and the initial risk value further inclu

Assignees

• Advanced New Technologies Co Ltd

Inventors

• Li Na

Classifications

• G06Q40/03

  Credit; Loans; Processing thereof · CPC title

• G06F16/9024

  Graphs; Linked lists (G06F16/9027 takes precedence) · CPC title

• G06Q40/08

  Insurance · CPC title

• G06Q30/018

  Certifying business or products · CPC title

• G06F2101/10

  Logarithmic or exponential functions · CPC title