Account authentication using synthetic merchants

1 . A computing device comprising:

one or more processors; and

memory storing instructions that, when executed by the one or more processors, cause the computing device to:

receive, from a merchants database, a plurality of different real merchant names;

process the plurality of different real merchant names to determine one or more name elements;

receive, from a user device, a request for access to an account associated with a user;

generate, based on the one or more name elements and based on a location associated with the account, one or more synthetic merchant names by:

providing, as input to a trained machine learning model, the one or more synthetic merchant names, wherein the trained machine learning model is trained to predict a believability of merchant names; and

receiving, as output from the trained machine learning model, a predicted believability of the one or more synthetic merchant names;

generate, based on the one or more synthetic merchant names, synthetic transaction data that indicates a financial transaction conducted by the account and associated with at least one of the one or more synthetic merchant names;

generate, based on the synthetic transaction data, a synthetic authentication question; and

provide the user device access to the account based on a response to the synthetic authentication question.

2 . The computing device of claim 1 , wherein the instructions, when executed by the one or more processors, cause the computing device to generate the one or more synthetic merchant names by causing the computing device to:

identify the location associated with the account; and

select, based on the location, at least one of the one or more name elements that corresponds to the location.

3 . The computing device of claim 1 , wherein the instructions, when executed by the one or more processors, cause the computing device to generate the one or more synthetic merchant names by causing the computing device to:

select a first name element of the one or more name elements that corresponds to a type of cuisine; and

select, based on the type of cuisine, a second name element of the one or more name elements that corresponds to a first name or surname.

4 . The computing device of claim 1 , wherein the trained machine learning model is trained to predict the believability of merchant names based on tagged training data comprising the plurality of different real merchant names.

5 . The computing device of claim 1 , wherein the instructions, when executed by the one or more processors, cause the computing device to generate the one or more synthetic merchant names based on comparing the one or more synthetic merchant names to the plurality of different real merchant names.

6 . The computing device of claim 1 , wherein the instructions, when executed by the one or more processors, cause the computing device to process the plurality of different real merchant names to determine the one or more name elements by causing the computing device to:

train a machine learning model to identify name elements by providing the machine learning model tagged data comprising a first portion of the plurality of different real merchant names;

provide, as input to the trained machine learning model, a second portion of the plurality of different real merchant names; and

receive, as output from the trained machine learning model, at least a portion of the one or more name elements.

7 . The computing device of claim 1 , wherein the one or more name elements comprise one or more of:

first names;

surnames;

geographical references; or

indications of goods or services.

8 . A method comprising:

receiving, by a computing device and from a merchants database, a plurality of different real merchant names;

processing, by the computing device, the plurality of different real merchant names to determine one or more name elements;

receiving, by the computing device and from a user device, a request for access to an account associated with a user;

generating, by the computing device, based on a location associated with the account, and based on the one or more name elements, one or more synthetic merchant names by:

providing, as input to a trained machine learning model, the one or more synthetic merchant names, wherein the trained machine learning model is trained to predict a believability of merchant names; and

receiving, as output from the trained machine learning model, a predicted believability of the one or more synthetic merchant names;

generating, by the computing device and based on the one or more synthetic merchant names, synthetic transaction data that indicates a financial transaction conducted by the account and associated with at least one of the one or more synthetic merchant names;

generating, by the computing device and based on the synthetic transaction data, a synthetic authentication question; and

providing the user device access to the account based on a response to the synthetic authentication question.

9 . The method of claim 8 , wherein generating the one or more synthetic merchant names comprises:

identifying the location associated with the account; and

selecting, based on the location, at least one of the one or more name elements that corresponds to the location.

10 . The method of claim 8 , wherein generating the one or more synthetic merchant names comprises:

selecting a first name element of the one or more name elements that corresponds to a type of cuisine; and

selecting, based on the type of cuisine, a second name element of the one or more name elements that corresponds to a first name or surname.

11 . The method of claim 8 , wherein the trained machine learning model is trained to predict the believability of merchant names based on tagged training data comprising the plurality of different real merchant names.

12 . The method of claim 8 , wherein generating the one or more synthetic merchant names is based on comparing the one or more synthetic merchant names to the plurality of different real merchant names.

13 . The method of claim 8 , wherein processing the plurality of different real merchant names to determine the one or more name elements comprises:

training a machine learning model to identify name elements by providing the machine learning model tagged data comprising a first portion of the plurality of different real merchant names;

providing, as input to the trained machine learning model, a second portion of the plurality of different real merchant names; and

receiving, as output from the trained machine learning model, at least a portion of the one or more name elements.

14 . The method of claim 8 , wherein the one or more name elements comprise one or more of:

first names;

surnames;

geographical references; or indications of goods or services.

15 . One or more non-transitory computer-readable media storing instructions that, when executed by one or more processors, cause a computing device to:

receive, from a merchants database, a plurality of different real merchant names;

process the plurality of different real merchant names to determine one or more name elements;

receive, from a user device, a request for access to an account associated with a user;

generate, based on the one or more name elements and based on a location associated with the account, one or more synthetic merchant names by:

providing, as input to a trained machine learning model, the one or more synthetic merchant names, wherein the trained machine learning model is trained to predict a believability of merchant names; and

receiving, as output from the trained machine learning model, a predicted believability of the one or more synthetic merchant names;

generate, based on the one or more synthetic merchant names, synthetic transaction data that indicates a financial transaction conducted by the account and associated with at least one of the one or more synthetic merchant names;

generate, based on the synthetic transaction data, a synthetic authentication question; and

provide the user device access to the account based on a response to the synthetic authentication question.

16 . The non-transitory computer-readable media of claim 15 , wherein the instructions, when executed by the one or more processors, cause the computing device to generate the one or more synthetic merchant names by causing the computing device to:

identify the location associated with the account; and

select, based on the location, at least one of the one or more name elements that corresponds to the location.

17 . The non-transitory computer-readable media of claim 15 , wherein the instructions, when executed by the one or more processors, cause the computing device to generate the one or more synthetic merchant names by causing the computing device to:

select a first name element of the one or more name elements that corresponds to a type of cuisine; and

select, based on the type of cuisine, a second name element of the one or more name elements that corresponds to a first name or surname.

18 . The non-transitory computer-readable media of claim 15 , wherein the trained machine learning model is trained to predict the believability of merchant names based on tagged training data comprising the plurality of different real merchant names.

19 . The non-transitory computer-readable media of claim 15 , wherein the instructions, when executed by the one or more processors, cause the computing device to generate the one or more synthetic merchant names based on comparing the one or more synthetic merchant names to the plurality of different real merchant names.

20 . The non-transitory computer-readable media of claim 15 , wherein the instructions, when executed by the one or more processors, cause the computing device to process the plurality of different real merchant names to determine the one or more name elements by causing the computing device to:

train a machine learning model to identify name elements by providing the machine learning model tagged data comprising a first portion of the plurality of different real merchant names;

provide, as input to the trained machine learning model, a second portion of the plurality of different real merchant names; and

receive, as output from the trained machine learning model, at least a portion of the one or more name elements.