ESTIMATING CHANGES TO USER RISK INDICATORS BASED ON MODELING OF SIMILARLY CATEGORIZED USERS
A data processing system communicates with a secure third-party database to obtain information about a plurality of users and generates a model usable to identify other users with similar characteristics. A scoring algorithm may be selected for use on user data based on the associated users identified with the model. The scoring algorithm determines an estimated score change for the user, and may provide the user information regarding how the user can achieve the estimated score change.
1 . A computerized method, performed by a computing system having one or more hardware computer processors and one or more non-transitory computer readable storage device storing software instructions executable by the computing system to perform the computerized method comprising:
accessing a score impact rule indicating one or more account types that impacts risk scores of users;
identifying an API token associated with a secured third-party risk database;
transmitting the API token and a request for risk data of a user to the third-party risk database;
accessing, via a secure communication channel established with the third-party risk database, risk data of the user;
identifying a plurality of groupings of data items included in the risk data of the user, each of the groupings including a plurality of data items associated with an entity;
determining, for each of the groupings of data items, an account type of a plurality of account types;
identifying a first account type of the plurality of account types that impacts risk scores when added to risk data of respective users and is not included in the determined account types associated with the user;
execute a score impact algorithm, based at least on the first account type and the risk data of the user, to determine a possible risk score change if an account of the first account type is added to the risk data of the user; and
providing, to the user, an indication of the possible risk score change.
2 . The computerized method of claim 1 , wherein the possible risk score changes indicates an estimated risk score.
3 . The computerized method of claim 1 , wherein the possible risk score changes indicates an estimated risk score increase or decrease.
4 . The computerized method of claim 1 , further comprising:
providing, to the user, an option to initiate addition of an account of the first account type to risk data of the user.
5 . The computerized method of claim 1 , wherein a first plurality of data items associated with a second entity indicate transactions of the user in a second account of the second account type.
6 . The computerized method of claim 1 , wherein the score impact algorithm is developed based at least on historical risk score changes of a plurality of other users in response to addition of respective accounts of the first account type to respective risk data of the users.
7 . The computerized method of claim 1 , further comprising:
executing a second score impact algorithm configured to estimate a second possible risk score associated with addition of a second account of a second account type to the risk data of the user; and
providing, to the user, a second indication of the second possible risk score change.
8 . The computerized method of claim 1 , further comprising:
receiving, via network communication with a user computing device,
selection of a third-party entity from a plurality of third-party entities indicated in a user interface displayed on the user computing device;
credentials for directly accessing, by proxy on behalf of the user via an application programming interface (API), a plurality of data items associated with the user stored in one or more databases associated with the selected third-party entity;
transmitting at least an API token associated with the selected third-party and the credential to one or more databases associated with the selected third-party entity; and
accessing a plurality of data items associated with the user, via an API communication channel established with the one or more databases associated with the selected third-party entity.
9 . The computerized method of claim 8 , wherein said identifying a plurality of groupings of data items further comprises:
selecting a first data item of the plurality of data items;
determining a recipient identified in the first data item;
identifying a grouping of data items each indicating the determined recipient, wherein the grouping of data items includes at least the first data item and one or more other data items;
determining, based at least on the identified grouping of data items, account data associated with an account of the user associated with the recipient, the account data comprising at least one or more of:
a number of data items each having time stamps within a predetermined time period;
average number of days between time stamps of sequential data items;
applying a first account identification rule, associated with the first account type, to the account data; and
determine, based on said application of the first account identification rule, a first confidence level indicating likelihood that the account is the first type of account.
10 . The computerized method of claim 9 , further comprising:
receiving, from the user computing device, confirmation that the account of the user is the first account type;
receiving, from the user computing device, a request to add the account with the recipient to risk data of the user at the secured third-party risk database;
generating, based on portions of the plurality of data items and the account data associated with the recipient, an account creation data package, the account creation data package formatted for ingestion at the secured third-party risk database to initiate addition of the account to risk data of the user;
identifying a security key associated with the secured third-party risk database; and
transmitting the security key and the account creation data package to the third-party risk database via a secure communication channel established with the third-party risk database.
11 . The computerized method of claim 10 , further comprising:
in response to determining that the first confidence level is above a first threshold, applying a first account scoring model to the account data, the first account scoring model configured to determine an expected change to a current risk score associated with the user;
requesting execution of a risk scoring algorithm using risk data of the user at the secured third-party risk database, wherein the risk scoring algorithm is based at least partly on portions of the plurality of data items or the account data included in the risk data of the user; and
providing risk score change information to the user computing device.
12 . A computing system comprising:
a hardware computer processor configured to perform operations comprising:
accessing a score impact rule indicating one or more account types that impacts risk scores of users;
identifying an API token associated with a secured third-party risk database;
transmitting the API token and a request for risk data of a user to the third-party risk database;
accessing, via a secure communication channel established with the third-party risk database, risk data of the user;
identifying a plurality of groupings of data items included in the risk data of the user, each of the groupings including a plurality of data items associated with an entity;
determining, for each of the groupings of data items, an account type of a plurality of account types;
identifying a first account type of the plurality of account types that impacts risk scores when added to risk data of respective users and is not included in the determined account types associated with the user;
executing a score impact algorithm, based at least on the first account type and the risk data of the user, to determine a possible risk score change if an account of the first account type is added to the risk data of the user; and
providing, to the user, an indication of the possible risk score change.
13 . The computing system of claim 12 , wherein the possible risk score changes indicates an estimated risk score.
14 . The computing system of claim 12 , wherein the possible risk score changes indicates an estimated risk score increase or decrease.
15 . The computing system of claim 12 , wherein the operations further comprise:
providing, to the user, an option to initiate addition of an account of the first account type to risk data of the user.
16 . The computing system of claim 12 , wherein a first plurality of data items associated with a second entity indicate transactions of the user in a second account of the second account type.
17 . The computing system of claim 12 , wherein the score impact algorithm is developed based at least on historical risk score changes of a plurality of other users in response to addition of respective accounts of the first account type to respective risk data of the users.
18 . The computing system of claim 12 , wherein the operations further comprise:
executing a second score impact algorithm configured to estimate a second possible risk score associated with addition of a second account of a second account type to the risk data of the user; and
providing, to the user, a second indication of the second possible risk score change.
19 . A non-transitory computer readable medium having processor-executable instructions stored thereon, the processor-executable instructions executable by a hardware computer processor to perform operations comprising:
accessing a score impact rule indicating one or more account types that impacts risk scores of users;
identifying an API token associated with a secured third-party risk database;
transmitting the API token and a request for risk data of a user to the third-party risk database;
accessing, via a secure communication channel established with the third-party risk database, risk data of the user;
identifying a plurality of groupings of data items included in the risk data of the user, each of the groupings including a plurality of data items associated with an entity;
determining, for each of the groupings of data items, an account type of a plurality of account types;
identifying a first account type of the plurality of account types that impacts risk scores when added to risk data of respective users and is not included in the determined account types associated with the user;
executing a score impact algorithm, based at least on the first account type and the risk data of the user, to determine a possible risk score change if an account of the first account type is added to the risk data of the user; and
providing, to the user, an indication of the possible risk score change.