Multi-factor authentication using location data

1 . A system for multi-factor authentication using location data, the system comprising:

one or more memories; and

one or more processors, coupled to the one or more memories, configured to:

obtain first location data, of a first level of granularity, associated with one or more exchanges based on obtaining records that are associated with one or more transactions related to an account;

obtain information from an entity location database that includes second location data associated with the one or more exchanges, wherein the second location data is associated with a second level of granularity that is higher than the first level of granularity and includes information of a physical location associated with one or more entities and identifies respective physical locations associated with the one or more exchanges,

wherein the entity location database includes information obtained via web scraping for the second location data;

pre-process entries of the entity location database based on searching using at least a merchant identifier, city, state, and zip code indicated by the records, to identify candidate entries likely associated with the one or more exchanges, thereby reducing computing overhead;

train, using the pre-processed entries of the entity location database, a machine learning model to determine a likelihood score that the one or more exchanges correspond to the respective physical locations;

provide information included in an entry of the entity location database and exchange information associated with an exchange, of the one or more exchanges, as inputs to the trained machine learning model, wherein the entry is associated with a physical location of the respective physical locations, that is associated with the one or more exchanges;

obtain, using the trained machine learning model, a score indicating a likelihood that the exchange corresponds to the physical location;

determine, based on the likelihood score satisfying a threshold, that the exchange is associated with the physical location;

receive, from a first device and over a network between the first device and the system, an access attempt associated with a request for access to the account that includes one or more credentials associated with a first layer of security to the account, wherein the first layer of security includes biometric information obtained via the first device;

authenticate the access attempt based on the one or more credentials based on the first layer of security;

detect an authentication event associated with granting access to the account, wherein the authentication event occurs after authenticating the access attempt is authenticated, for performing additional operations associated with a second layer of security related to the account;

provide, based on detecting the authentication event, one or more authentication queries that are based on the second location data,

wherein the one or more authentication queries are associated with at least one of the respective physical locations associated with the one or more exchanges, and

wherein the one or more authentication queries are provided to the first device in a prompt displayed on the first device;

grant or deny access to the account based on one or more responses to the one or more authentication queries;

re-train the machine learning model based on feedback information that is provided to the machine learning model via a feedback loop,

wherein the feedback information is an output of the machine learning model including an indication of whether a predicted match between an entry in the entity location database and the exchange is correct; and

utilize the re-trained machine learning model in subsequent likelihood score determinations.

2 . The system of claim 1 , wherein an authentication query, of the one or more authentication queries, is associated with requesting confirmation of a location at which an exchange, of the one or more exchanges, occurred, and wherein the authentication query identifies:

exchange data associated with the exchange, wherein the exchange data is indicated by the records, and

an indication of one or more physical locations, where a physical location, of the one or more physical locations, is the location at which the exchange occurred as identified by the respective physical locations associated with the one or more exchanges.

3 . The system of claim 1 , wherein the first location data is associated with a first level of granularity and the second location data is associated with a second level of granularity, and wherein the second level of granularity is a higher level of granularity than the first location data.

4 . The system of claim 1 , wherein the first location data includes at least one of:

state information,

city information, or

zip code information.

5 . The system of claim 4 , wherein the second location data includes address information.

6 . The system of claim 1 , wherein an authentication query, of the one or more authentication queries, is associated with requesting confirmation of a payment technique associated with an exchange of the one or more exchanges, wherein the payment technique is indicated by the records.

7 . The system of claim 1 , wherein the machine learning model is re-trained based on the feedback information related to whether a predicted match between the one or more exchanges and the respective physical locations is correct.

8 . A method of multi-factor authentication using location data, comprising:

obtaining, by a device, first location data associated with one or more exchanges based on obtaining records that are associated with one or more transactions related to an account, wherein the first location data is associated with a first level of granularity;

obtaining, by the device and from an entity location database, information that includes second location data associated with the one or more exchanges, wherein the second location data is associated with a second level of granularity that is higher than the first level of granularity and includes information of a physical location associated with one or more entities;

pre-process entries of the entity location database based on searching using at least a merchant identifier, city, state, and zip code indicated by the records, to identify candidate entries likely associated with the one or more exchanges, thereby reducing computing overhead;

training, by the device and using the pre-processed entries, a machine learning model to determine a score indicating a likelihood that an exchange of the one or more exchanges corresponds to a particular second location data;

obtaining, by the device and using the trained machine learning model, the score;

determining, by the device and based on the score satisfying a threshold, that the exchange is associated with the particular second location data;

receiving, by the device, from another device, and over a network between the device and the other device, an access attempt associated with a request for access to the account, that includes one or more credentials associated with a first layer of security to the account, wherein the first layer of security includes biometric information obtained via the other device;

authenticating, by the device, the access attempt based on the one or more credentials based on the first layer of security;

detecting, by the device, an authentication event associated with granting access to the account, wherein the authentication event occurs after authenticating the access attempt is authenticated, for performing additional operations associated with a second layer of security related to the account;

providing, by the device and based on detecting the authentication event, one or more authentication queries that are based on the particular second location data,

wherein the one or more authentication queries are provided to the device in a prompt displayed on the device;

granting or denying access to the account based on one or more responses to the one or more authentication queries;

re-training the machine learning model based on feedback information that is provided to the machine learning model via a feedback loop,

wherein the feedback information is an output of the machine learning model including an indication of whether a predicted match between an entry in the entity location database and the exchange is correct; and

utilizing the re-trained machine learning model in subsequent score determinations.

9 . The method of claim 8 , wherein the second location data includes address information associated with respective entity locations at which the one or more exchanges occurred.

10 . The method of claim 8 , wherein detecting the authentication event comprises:

obtaining credentials associated with the account;

authenticating the credentials associated with the account; and

detecting the authentication event based on at least one of a purpose associated with the access attempt or device information associated with the access attempt.

11 . The method of claim 8 , wherein authenticating the access attempt to the account comprises:

obtaining a first response, of the one or more responses, associated with a first authentication query of the one or more authentication queries, wherein the first response is incorrect;

providing a second one or more authentication queries of the one or more authentication queries based on the first response being incorrect;

authenticating the access attempt based on a second one or more responses to the second one or more authentication queries being correct; and

performing an action associated with a fraud detection model that is associated with the account based on the first response being incorrect.

12 . The method of claim 8 , wherein the first location data is included in information provided by a backend device as part of completing an exchange, of the one or more exchanges, that is associated with an entity at which the exchange occurred.

13 . The method of claim 8 , wherein a first authentication query, of the one or more authentication queries, requests a confirmation of an address associated with an exchange, of the one or more exchanges, wherein the address is indicated by the second location data, and

wherein a second authentication query, of the one or more authentication queries, requests a confirmation of a method of payment, indicated by the records, associated with the exchange.

14 . The method of claim 8 , wherein determining the second location data comprises:

searching, using the first location data, the entity location database;

identifying an entry in the entity location database that is associated with an exchange of the one or more exchanges based on the entry including information that is associated with the first location data; and

determining the second location data based on information included in the entry of the entity location database.

15 . The method of claim 8 , wherein authenticating the access attempt to the account comprises:

determining whether the one or more responses to the one or more authentication queries are correct; and

granting access to the account if the one or more responses are correct; or

denying access to the account if the one or more responses are incorrect.

16 . A non-transitory computer-readable medium storing a set of instructions, the set of instructions comprising:

one or more instructions that, when executed by one or more processors of a device, cause the device to:

obtain records associated with one or more exchanges that are associated with an account, wherein the records include first location data, of a first level of granularity, associated with the one or more exchanges;

obtain, from an entity location database, information that includes second location data associated with the one or more exchanges based on the records, wherein the second location data includes address information associated with respective entity locations at which the one or more exchanges occurred;

pre-process entries of the entity location database based on searching using at least a merchant identifier, city, state, and zip code indicated by the records, to identify candidate entries likely associated with the one or more exchanges, thereby reducing computing overheard;

train, using the pre-processed entries, a machine learning model to determine a score indicating a likelihood that an exchange of the one or more exchanges is related to a particular second location data;

obtain, using the trained machine learning model, the score;

determine that the exchange is associated with a particular entity location based on the score satisfying a threshold;

receive, from another device and over a network between the device and the other device, an access attempt associated with a request for access to the account, that includes one or more credentials associated with a first layer of security to the account, wherein the first layer of security includes biometric information obtained via the other device;

authenticate the access attempt based on the one or more credentials based on the first layer of security;

detect an authentication event associated with an access attempt to the account, wherein the authentication event occurs after authenticating the access attempt is authenticated, for performing additional operations associated with a second layer of security related to the account;

provide, based on detecting the authentication event, one or more authentication queries that are based on the particular entity location,

wherein the one or more authentication queries are provided to the other device in a prompt displayed on the other device;

grant or deny access to the account based on one or more responses to the one or more authentication queries;

re-train the machine learning model based on feedback information that is provided to the machine learning model via a feedback loop,

wherein the feedback information is an output of the machine learning model including an indication of whether a predicted match between an entry in the entity location database and the exchange is correct; and

utilize the re-trained machine learning model in subsequent score determinations.

17 . The non-transitory computer-readable medium of claim 16 , wherein the one or more instructions, that cause the device to detect the authentication event, cause the device to:

obtain the credentials associated with the account;

authenticate the credentials associated with the account; and

detect the authentication event based on at least one of a purpose associated with the access attempt or device information associated with the access attempt.

18 . The non-transitory computer-readable medium of claim 16 , wherein the one or more instructions, that cause the device to authenticate the access attempt to the account, cause the device to:

obtain a first response, of the one or more responses, associated with a first authentication query of the one or more authentication queries, wherein the first response is incorrect;

provide a second one or more authentication queries of the one or more authentication queries based on the first response being incorrect; and

authenticate the access attempt based on a second one or more responses to the second one or more authentication queries being correct.

19 . The non-transitory computer-readable medium of claim 18 , wherein the one or more instructions further cause the device to:

perform an action associated with a fraud detection model that is associated with the account based on the first response being incorrect, wherein the action causes a score or parameter indicating a likelihood of fraud associated with the account to be increased.

20 . The non-transitory computer-readable medium of claim 16 , wherein the one or more instructions, that cause the device to determine the second location data, cause the device to:

search, using the first location data, the entity location database;

identify an entry in the entity location database that is associated with an exchange of the one or more exchanges; and

determine an address associated with the exchange based on information included in the entry of the entity location database.