Systems and methods for early account score and notification

• US 8,725,613 B1
• Filed: 04/26/2011
• Issued: 05/13/2014
• Est. Priority Date: 04/27/2010
• Status: Active Grant

• Current Assignee: Experian Information Solutions Incorporated (Experian plc)
• Original Assignee: Experian Information Solutions Incorporated (Experian plc)
• Inventors: Celka, Christopher J., Rojas, Cristian R.
• Primary Examiner(s): Poinvil, Frantzy
• Application Number:
  US13/094,497
• Time in Patent Office: 1,113 Days
• Field of Search: 705 4- 44
• US Class Current: 705/35
• CPC Class Codes: G06Q 40/03 Credit; Loans; Processing thereof

Abstract

The embodiments illustrated herein provide systems and methods for predicting or identifying early life account fraud at a point-in-time within existing account portfolios. The identity-level linking of inquiry data described herein allow for detecting inconsistent and/or suspicious use of identity data elements across multiple applications that traditional fraud detection systems may consider to be different consumers. The system can be configured to calculate or generate a fraud score after a credit account is booked and/or opened by a financial institution or other third party. Even though the likelihood of fraud is assessed at the time when an applicant applies for credit, there may still be some applicants that successfully pass through authentication and/or fraud tools.

Claims

1. A computer-implemented method for fraud monitoring, the computer-implemented method comprising:

accessing consumer data related to a consumer, the consumer data comprising a first identity data element, a second identity data element, and a third identity data element;

receiving, by the computer processor, a set of credit inquiry records, wherein the set of credit inquiry records comprises requests received from third parties for credit bureau data for the consumer within a predetermined time period, each of the credit inquiry records of the subset comprising a first identity data element, a second identity data element and a third identity data element;

associating the consumer with at least one of the records of the credit inquiry records based on the first identity data element of the consumer being the same or similar to the first identity data element of the at least one credit inquiry record;

for the credit inquiry records associated with the consumer,

comparing the second identity data element of the consumer to each of the second identity data elements of the associated credit inquiry records;

comparing the third identity data element of the consumer to each of the third identity data elements of the associated credit inquiry records;

calculating a first set of attributes for the consumer based on the comparisons of the second identity data elements and the third identity data elements, wherein the first set of attributes includes one or more of;

number of times the same or similar first identity data element is used with a different second identity data element,

number of times the same or similar first identity element is used with a different third identity data element,

number of times the same or similar second identity data element with a different third identity data element,

number of times the same or similar third identity data element with a different second identity data element,

number of different second data elements in a given time period, or

number of different third data elements in a given time period;

applying by the computer processor a fraud model using the first set of calculated attributes to predict a likelihood of fraud;

generating by the computer processor a first fraud score based on the applying the fraud model using the first set of calculated attributes;

associating the consumer with at least one of the records of the credit inquiry records based on the second identity data element of the consumer being the same or similar to the second identity data element of the at least one credit inquiry record;

for the credit inquiry records associated with the consumer,

comparing the first identity data element of the consumer to each of the first identity data elements of the associated credit inquiry records;

comparing the third identity data element of the consumer to each of the third identity data elements of the associated credit inquiry records;

calculating a second set of attributes for the consumer based on the comparisons of the first identity data elements and the third identity data elements, wherein the second set of attributes includes one or more of;

number of times the same or similar first identity element is used with a different third identity data element,

number of times the same or similar second identity data element is used with a different first identity data element,

number of times the same or similar second identity data element with a different third identity data element,

number of times the same or similar third identity data element is used with a different first identity data element,

number of different first identity data elements in a given time period,

number of different third data elements in a given time period;

applying by the computer processor a fraud model using the second set of calculated attributes to predict a likelihood of fraud;

generating by the computer processor a second fraud score based on the applying the fraud model using the second set of calculated attributes; and

calculating an overall fraud score using the first fraud score and the second fraud score;

outputting the overall fraud score for the consumer.

2. The computer-implemented method for fraud monitoring of claim 1, wherein each of the first, second and third identity data elements comprises one of a name of the consumer, a social security number of the consumer, a telephone number of the consumer, or an address of the consumer.

3. The computer-implemented method for fraud monitoring of claim 1, wherein calculating the overall fraud score further comprises using a number of credit inquiry records associated with the consumer.

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

receiving, by the computer processor, a set of authentication inquiry records, wherein the set of authentication inquiry records comprises inquiries received from third parties to a credit bureau received within a predetermined time period, and wherein each inquiry is a request to verify and/or authenticate an identity or identification information of the consumer, each of the authentication inquiry records of the subset comprising a first identity data element, a second identity data element and a third identity data element;

associating the consumer with at least one of the records of the authentication inquiry records based on the first identity data element of the consumer being the same or similar to the first identity data element of the at least one authentication inquiry record;

for the authentication inquiry records associated with the consumer,

comparing the second identity data element of the consumer to each of the second identity data elements of the associated authentication inquiry records;

comparing the third identity data element of the consumer to each of the third identity data elements of the associated authentication inquiry records;

calculating the first set of attributes for the consumer based on the comparisons of the second identity data elements and the third identity data elements, wherein the first set of attributes includes one or more of;

number of times the same or similar first identity data element is used with a different second identity data element,

number of times the same or similar first identity element is used with a different third identity data element,

number of times the same or similar second identity data element with a different third identity data element,

number of times the same or similar third identity data element with a different second identity data element,

number of different second data elements in a given time period, or

number of different third data elements in a given time period;

applying by the computer processor the fraud model using the third set of calculated attributes to predict a likelihood of fraud;

generating by the computer processor the first fraud score based on the applying the fraud model using the first set of calculated attributes;

associating the consumer with at least one of the records of the authentication inquiry records based on the second identity data element of the consumer being the same or similar to the second identity data element of the at least one authentication inquiry record;

for the authentication inquiry records associated with the consumer,

comparing the first identity data element of the consumer to each of the first identity data elements of the associated authentication inquiry records;

comparing the third identity data element of the consumer to each of the third identity data elements of the associated authentication inquiry records;

calculating the second set of attributes for the consumer based on the comparisons of the first identity data elements and the third identity data elements, wherein the second set of attributes includes one or more of;

number of times the same or similar first identity element is used with a different third identity data element,

number of times the same or similar second identity data element is used with a different first identity data element,

number of times the same or similar second identity data element with a different third identity data element,

number of times the same or similar third identity data element is used with a different first identity data element,

number of different first identity data elements in a given time period, number of different third data elements in a given time period;

applying by the computer processor the fraud model using the fourth set of calculated attributes to predict a likelihood of fraud;

generating by the computer processor the second fraud score based on the applying the fraud model using the second set of calculated attributes.

5. A computer system for fraud monitoring, the computer system comprising:

a computer processor configured to execute modules comprising at least;

a data access module configured to access consumer data related to a consumer, the consumer data comprising a first identity data element, a second identity data element and a third identity data element;

an inquiry record access module configured to receive a set of credit inquiry records, wherein the set of credit inquiry records comprises requests received from third parties for credit bureau data for the consumer within a predetermined time period, each of the credit inquiry records of the subset comprising a first identity data element, a second identity data element and a third identity data element;

a first association module configured to associate the consumer with at least one of the records of the credit inquiry records based on the first identity data element of the consumer being the same or similar to the first identity data element of the at least one credit inquiry record;

a first scoring module configured to calculate a first fraud score based on, for the credit inquiry records associated with the consumer,

comparing the second identity data element of the consumer to each of the second identity data elements of the associated credit inquiry records;

comparing the third identity data element of the consumer to each of the third identity data elements of the associated credit inquiry records;

calculating a first set of attributes for the consumer based on the comparisons of the second identity data elements and the third identity data elements, wherein the first set of attributes includes one or more of;

number of times the same or similar first identity data element is used with a different second identity data element,

number of times the same or similar first identity element is used with a different third identity data element,

number of times the same or similar second identity data element with a different third identity data element,

number of times the same or similar third identity data element with a different second identity data element,

number of different second data elements in a given time period, or

number of different third data elements in a given time period;

applying by the computer processor a fraud model using the first set of calculated attributes to predict a likelihood of fraud;

generating by the computer processor a first fraud score based on the applying the fraud model using the first set of calculated attributes;

a second association module configured to associate the consumer with at least one of the records of the credit inquiry records based on the second identity data element of the consumer being the same or similar to the second identity data element of the at least one credit inquiry record;

a second scoring module configured to calculate a second fraud score based on, for the credit inquiry records associated with the consumer,

comparing the first identity data element of the consumer to each of the first identity data elements of the associated credit inquiry records;

comparing the third identity data element of the consumer to each of the third identity data elements of the associated credit inquiry records;

calculating a second set of attributes for the consumer based on the comparisons of the first identity data elements and the third identity data elements, wherein the second set of attributes includes one or more of;

number of times the same or similar first identity element is used with a different third identity data element,

number of times the same or similar second identity data element is used with a different first identity data element,

number of times the same or similar second identity data element with a different third identity data element,

number of times the same or similar third identity data element is used with a different first identity data element,

number of different first identity data elements in a given time period,

number of different third data elements in a given time period;

applying by the computer processor a fraud model using the second set of calculated attributes to predict a likelihood of fraud;

generating by the computer processor a second fraud score based on the applying the fraud model using the second set of calculated attributes; and

an overall scoring module configured to calculate an overall fraud score using the first fraud score and the second fraud score; and

an output module configured to output the overall fraud score.

6. The computer system for fraud monitoring of claim 5, wherein each of the plurality of first, second and third identity data elements comprises one of a name of the consumer, a social security number of the consumer, a telephone number of the consumer, or an address of the consumer.

7. The computer system for fraud monitoring of claim 5, wherein the overall fraud score is further based on a number of inquiry records associated with the consumer.

8. The computer system of claim 5, wherein the inquiry record access module is further configured to receive a set of authentication inquiry records,

wherein the set of authentication inquiry records comprises inquiries received from third parties to a credit bureau received within a predetermined time period, and wherein each inquiry is a request to verify and/or authenticate an identity or identification information of the consumer, each of the authentication inquiry records of the subset comprising a first identity data element, a second identity data element and a third identity data element;

wherein the first association module is further configured to associate the consumer with at least one of the records of the authentication inquiry records based on the first identity data element of the consumer being the same or similar to the first identity data element of the at least one authentication inquiry record;

wherein the first scoring module is further configured to calculate the first fraud score based on, for the authentication inquiry records associated with the consumer,

comparing the second identity data element of the consumer to each of the second identity data elements of the associated authentication inquiry records;

comparing the third identity data element of the consumer to each of the third identity data elements of the associated authentication inquiry records;

calculating the first set of attributes for the consumer based on the comparisons of the second identity data elements and the third identity data elements, wherein the first set of attributes includes one or more of;

number of times the same or similar first identity data element is used with a different second identity data element,

number of times the same or similar first identity element is used with a different third identity data element,

number of times the same or similar second identity data element with a different third identity data element,

number of times the same or similar third identity data element with a different second identity data element,

number of different second data elements in a given time period, or

number of different third data elements in a given time period;

applying by the computer processor the fraud model using the first set of calculated attributes to predict a likelihood of fraud;

generating by the computer processor the first fraud score based on the applying the fraud model using the third set of calculated attributes;

wherein the second association module is further configured to associate the consumer with at least one of the records of the authentication inquiry records based on the second identity data element of the consumer being the same or similar to the second identity data element of the at least one authentication inquiry record;

wherein the second scoring module is further configured to calculate the second fraud score based on, for the authentication inquiry records associated with the consumer,

comparing the first identity data element of the consumer to each of the first identity data elements of the associated authentication inquiry records;

comparing the third identity data element of the consumer to each of the third identity data elements of the associated authentication inquiry records;

calculating the second set of attributes for the consumer based on the comparisons of the first identity data elements and the third identity data elements, wherein the second set of attributes includes one or more of;

number of times the same or similar first identity element is used with a different third identity data element,

number of times the same or similar second identity data element is used with a different first identity data element,

number of times the same or similar second identity data element with a different third identity data element,

number of times the same or similar third identity data element is used with a different first identity data element,

number of different first identity data elements in a given time period,

number of different third data elements in a given time period;

applying by the computer processor the fraud model using the second set of calculated attributes to predict a likelihood of fraud;

generating by the computer processor the second fraud score based on the applying the fraud model using the second set of calculated attributes.