Reducing false positives in entity matching based on image-linking graphs
Methods and systems are presented for performing comprehensive and accurate matching of user accounts with one or more known entities based on image-linking graphs. Images related to each known entity are retrieved from one or more online sources. Faces are extracted from the images. Based on attributes of the faces in the images, an image-linking graph is generated for the entity. When a user account is determined to be a potential match for the entity based on text-based attributes, an image associated with the account may be obtained. If the image matches with any one of the faces in the image-linking graph, an action is performed to the user account based on a position of the matched face in the image-linking graph.
1 . A system, comprising:
a non-transitory memory storing instructions; and
one or more hardware processors coupled with the non-transitory memory and configured to execute the instructions to cause the system to:
obtain an image of a first person associated with a user account;
access a graph associated with an entity, wherein the graph links a plurality of persons associated with the entity based on one or more images associated with the entity, and wherein two persons from the plurality of persons are linked in the graph when the two persons appear in at least one image of the one or more images associated with the entity;
determine a first position of the first person within the graph based on the image of the first person;
identify a second person who has a particular type of relationship with the entity based on a second position of the second person in the graph;
determine a level of connectedness between the first person and the second person in the graph; and
perform an action to the user account based on the first position of the first person within the graph and the level of connectedness between the first person and the second person.
2 . The system of claim 1 , wherein executing the instructions further causes the system to:
detect that the first person is in the one or more images.
3 . The system of claim 1 , wherein the entity corresponds to a particular type of entity, and wherein performing the action is further based on the particular type of entity and comprises removing access to one or more resources of the system from the user account.
4 . The system of claim 1 , wherein the entity corresponds to a particular type of entity, and wherein performing the action is further based on the particular type of entity and comprises providing the user account access to one or more resources of the system.
5 . The system of claim 1 , wherein executing the instructions further causes the system to:
identify one or more persons in the graph who are connected to the first person, wherein performing the action is further based on the one or more persons identified in the graph.
6 . The system of claim 1 , wherein the level of connectedness is determined based on at least one of a number of connections between the first person and the second person in the graph or a weight associated with each of the connections between the first person and the second person.
7 . The system of claim 1 , wherein the user account is a first user account, and wherein executing the instructions further causes the system to:
access transaction data associated with one or more transactions conducted between the first user account and a second user account of the second person, wherein performing the action is further based on the transaction data.
8 . A method comprising:
receiving, by a computer system, a transaction request associated with a first user account;
accessing, by the computer system, a graph associated with an entity, wherein the graph links a plurality of persons associated with the entity based on one or more images associated with the entity, and wherein two persons from the plurality of persons are linked in the graph when the two persons appear in at least one image of the one or more images associated with the entity;
determining, by the computer system, a first position of a first person associated with the first user account within the graph based on an image of the first person;
identifying, by the computer system, a second person who has a particular type of relationship with the entity based on a second position of the second person in the graph;
accessing, by the computer system, transaction data associated with transactions conducted between the first user account and a second user account of the second person; and
determining, by the computer system, whether to authorize the transaction request, deny the transaction request, or prompt the first person for information based on the first position of the first person and the transaction data.
9 . The method of claim 8 , further comprising:
determining a transaction pattern for the transactions conducted between the first user account and the second user account, wherein the determining whether to authorize the transaction request, deny the transaction request, or prompt the person for information is further based on the transaction pattern.
10 . The method of claim 8 , wherein the second person is a leader of the entity.
11 . The method of claim 8 , further comprising:
predicting that the first person is associated with the entity based on attributes associated with the first person.
12 . The method of claim 8 , wherein the one or more images are retrieved from a plurality of sources comprises at least one of a news outlet, a social media platform, or a website of an organization.
13 . The method of claim 8 , further comprising:
determining a relatedness between the first person and the second person in the graph, wherein the determining whether to authorize the transaction request, deny the transaction request, or prompt the first person for information is further based on the relatedness between the first person and the second person.
14 . The method of claim 13 , wherein the relatedness between the first person and the second person is determined based on at least one of a number of connections between the first person and the second person in the graph or a weight associated with each of the connections between the first person and the second person.
15 . A non-transitory machine-readable medium having stored thereon machine-readable instructions executable to cause a machine to perform operations comprising:
receiving, from a device of a first person, a request to conduct a transaction through a user account with a service provider;
obtaining an image of the first person;
accessing a graph associated with an entity, wherein the graph links a plurality of persons associated with the entity based on one or more images associated with the entity, and wherein two persons from the plurality of persons are linked in the graph when the two persons appear in at least one image of the one or more images associated with the entity;
determining a first position of the first person within the graph based on the image of the first person;
identifying a second person who has a particular type of relationship with the entity based on a second position of the second person in the graph;
determining a level of connectedness between the first person and the second person in the graph; and
determining how to process the transaction based on the first position of the first person in the graph and the level of connectedness between the first person and the second person.
16 . The non-transitory machine-readable medium of claim 15 , wherein the entity corresponds to a user of the user account.
17 . The non-transitory machine-readable medium of claim 15 , wherein the operations further comprise:
processing the transaction based on the determining how to process the transaction.
18 . The non-transitory machine-readable medium of claim 15 , wherein the operations further comprise:
identifying one or more persons in the graph who are connected to the first person, wherein the determining how to process the transaction is further based on the one or more persons connected to the first person in the graph.
19 . The non-transitory machine-readable medium of claim 15 , wherein the level of connectedness is determined based on at least one of a number of connections between the first person and the second person in the graph or a weight associated with each of the connections between the first person and the second person.
20 . The non-transitory machine-readable medium of claim 15 , wherein the user account is a first user account, and wherein the operations further comprise:
accessing transaction data associated with one or more transactions conducted between the first user account and a second user account of the second person, wherein the determining how to process the transaction is further based on the transaction data.