Systems and methods for detecting fraudulent users

1 . A computer-implemented method, comprising:

retrieving account information of a plurality of users comprising user information from a datastore;

generating a graph using the retrieved account information, wherein generating the graph further comprising:

generating a plurality of user nodes to represent each of the plurality of users;

generating a plurality of field nodes to represent a plurality of fields associated with the plurality of users, and wherein each of the plurality of field nodes includes a value for the respective field;

generating one or more edges between each user node and each field node associated with each user node;

identifying a node pair, wherein each node pair comprises a first field node with an edge to a first user node and a second field node with an edge to a second user node, wherein the first field node and the second field node represent the same type of field;

determining a similarity score between the first field node and the second field node of the node pair; and

connecting the first field node and the second field node via an edge if the similarity score is more than a pre-specified threshold; and

determining one or more cliques in the graph based on an edge limit defining at least one clique.

2 . The computer-implemented method of claim 1 , further comprising:

identifying one or more user nodes in the graph, wherein the one or more user nodes are a member of the determined one or more cliques; and

reporting the identified one or more user nodes for further review.

3 . The computer-implemented method of claim 2 , wherein reporting the identified one or more user nodes for further review further comprises preventing the users associated to the identified one or more user nodes from performing any transactions.

4 . The computer-implemented method of claim 1 , wherein the graph is an undirected graph.

5 . The computer-implemented method of claim 4 , wherein the graph is a disconnected graph.

6 . The computer-implemented method of claim 1 , wherein determining the similarity score further comprises:

determining a distance between the value of the first field node and the value of the second field node; and

calculating a similarity score based on the determined distance.

7 . The computer-implemented method of claim 6 , wherein the distance between the value of the first field node and the value of the second field node is an edit distance.

8 . A computer-implemented method, comprising:

receiving data relating to an entity, wherein the data is categorized according to a plurality of fields, wherein each of the plurality of fields associated with the entity is defined by one of a plurality of field types, and where each of the fields are respectively associated with a plurality of field values;

matching at least one of the field values to a value associated with a matching node in a graph database, wherein the graph database is defined by a plurality of nodes having one of the plurality of field types, and wherein the matching node has the same type as the field associated with the at least one of the field values;

searching the graph database for fuzzy matches for each of the plurality of field values relating to the entity by comparing each of the plurality of field values to a subset of the plurality of nodes having one of the plurality of field types matching the one of the plurality of field types associated with the respective one of the plurality of field values; and

based on finding one or more fuzzy matches, executing a fraud response action.

9 . The computer-implemented of claim 8 , wherein the fraud response action comprises generating a report including the one or more fuzzy matches and transmitting the report to one or more evaluators.

10 . The computer-implemented of claim 9 , wherein the one or more evaluators comprises one or more machine learning models configured to evaluate the report and determine a validity of the report.

11 . The computer-implemented of claim 8 , wherein the fraud response action further comprises preventing the entity from executing any action.

12 . The computer-implemented method of claim 8 , wherein the fraud response action further comprises automatically closing an account.

13 . The computer-implemented method of claim 8 , wherein each of the one or more fuzzy matches is assigned a value, and wherein executing the fraud response is based on an aggregate of the values exceeding a response threshold.

14 . A non-transitory computer-readable medium storing instructions that when executed by a processor effectuate operations comprising:

retrieving account information of a plurality of users comprising user information from a datastore;

generating a graph using the retrieved account information, wherein generating the graph further comprising:

generating a plurality of user nodes to represent each of the plurality of users;

generating a plurality of field nodes to represent a plurality of fields associated with the plurality of users, and wherein each of the plurality of field nodes includes a value for the respective field;

generating one or more edges between each user node and each field node associated with each user node;

identifying a node pair, wherein each node pair comprises a first field node with an edge to a first user node and a second field node with an edge to a second user node, wherein the first field node and the second field node represent the same type of field;

determining a similarity score between the first field node and the second field node of the node pair;

connecting the first field node and the second field node via an edge if the similarity score is more than a pre-specified threshold; and

determining one or more cliques in the graph.

15 . The non-transitory computer-readable medium of claim 14 , wherein the operations comprise:

identifying one or more user nodes in the graph, wherein the one or more user nodes are a member of the determined one or more cliques; and

reporting the identified one or more user nodes for further review.

16 . The non-transitory computer-readable medium of claim 15 , wherein reporting the identified one or more user nodes for further review further comprises preventing the users associated to the identified one or more user nodes from performing any transactions.

17 . The non-transitory computer-readable medium of claim 14 , wherein the graph is an undirected graph.

18 . The non-transitory computer-readable medium of claim 14 , wherein the graph is a disconnected graph.

19 . The non-transitory computer-readable medium of claim 14 , wherein the operations comprise:

determining a distance between the value of the first field node and the value of the second field node; and

calculating a similarity score based on the determined distance.

20 . The non-transitory computer-readable medium of claim 19 , wherein the distance between the value of the first field node and the value of the second field node is an edit distance.