Systems and methods for generating personalized lending scores

Patent metadata

Field	Value
Publication number	US-2018189872-A1
Application number	US-201715399318-A
Country	US
Kind code	A1
Filing date	Jan 5, 2017
Priority date	Jan 5, 2017
Publication date	Jul 5, 2018
Grant date	—

Abstract

A scoring engine computing device for generating personalized lending scores is provided. The scoring engine computing device receives a request including a cardholder identifier associated with a candidate cardholder, determines demographic data associated with the candidate cardholder, and retrieves transaction data for a plurality of cardholders including the candidate cardholder and a set of peer cardholders. Each cardholder of the set of peer cardholders is associated with the determined demographic data of the candidate cardholder, and the transaction data is associated with transactions for a plurality of spending categories. The scoring engine computing device further normalizes the transaction data associated with the candidate cardholder by category, generates a personalized lending score associated with the candidate cardholder that indicates a spending trend of the candidate cardholder, and transmits the personalized lending score to a requestor computing device.

First claim

1 . A scoring engine computing device including a processor in communication with a memory, said processor programmed to:
receive a request from a requestor computing device, the request including a cardholder identifier associated with a candidate cardholder; determine demographic data associated with the candidate cardholder based at least in part on the request; retrieve transaction data for a plurality of cardholders including the candidate cardholder and a set of peer cardholders, each cardholder of the set of peer cardholders associated with the determined demographic data of the candidate cardholder, wherein the transaction data is associated with transactions for a plurality of spending categories; normalize, for each spending category, the transaction data associated with the candidate cardholder based at least in part on the transaction data associated with the set of peer cardholders; generate a personalized lending score associated with the candidate cardholder based at least in part on the normalized transaction data, wherein the personalized lending score indicates a spending trend of the candidate cardholder; and transmit the personalized lending score to the requestor computing device.

2 . The scoring engine computing device of claim 1 , wherein said processor is further programmed to:
apply a plurality of predetermined ratings to the spending categories of the transaction data, wherein a negative rating indicates a spending category associated with an undesirable spending habit and a positive rating indicates a spending category associated with a desirable spending habit; and generate the personalized lending score based at least in part on the spending category of the transaction data and the predetermined ratings.

3 . The scoring engine computing device of claim 2 , wherein a net rating for all spending categories is equal to 1 .

4 . The scoring engine computing device of claim 1 , wherein the demographic data includes an age group of the cardholder, an income group of the cardholder, a geographical residence location of the cardholder, and combinations thereof.

5 . The scoring engine computing device of claim 1 , wherein the generated personalized lending score includes an expense rating wherein a negative expense rating indicates that the candidate cardholder's spending is dominated by undesirable spending categories and a positive expense rating indicates that the candidate cardholder's spending is dominated by desirable spending categories.

6 . The scoring engine computing device of claim 1 , wherein the generated personalized lending score includes a total expense rating wherein a negative total expense rating indicates that the candidate cardholder generally spends more than a peer cardholder associated with the demographic data, and a positive total expense rating indicates that the candidate cardholder generally saves more than a peer cardholder associated with the demographic data.

7 . The scoring engine computing device of claim 1 , wherein the plurality of spending categories includes one or more of the following spending categories: groceries, high end groceries, low end groceries, fast food restaurants, fine dining restaurants, healthy eating restaurants, travel, entertainment, gambling, adult entertainment, utilities, charity, preventive healthcare, and combinations thereof.

8 . The scoring engine computing device of claim 1 , wherein said processor is further programmed to:
generate a recommendation associated with the candidate cardholder by comparing the personalized lending score to at least one predetermined threshold, the recommendation recommending to approve or decline the candidate cardholder for a loan; and transmit the recommendation with the personalized lending score to the requestor computing device, wherein a lending party associated with the requestor computing device approves or declines the candidate cardholder for the load based at least in part on the recommendation.

9 . The scoring engine computing device of claim 1 , wherein the personalized lending score is generated based on additional third party information associated with the candidate cardholder including bank account balance, bank account assets, credit report information, credit score, social media score, and combinations thereof.

10 . A method for generating a personalized lending score associated with a candidate cardholder, said method performed using a scoring engine computing device including a processor in communication with a memory, said method comprising:
receiving a request from a requestor computing device, the request including a cardholder identifier associated with a candidate cardholder; determining demographic data associated with the candidate cardholder based at least in part on the request; retrieving transaction data for a plurality of cardholders including the candidate cardholder and a set of peer cardholders, each cardholder of the set of peer cardholders associated with the determined demographic data of the candidate cardholder, wherein the transaction data is associated with transactions for a plurality of spending categories; normalizing, for each spending category, the transaction data associated with the candidate cardholder based at least in part on the transaction data associated with the set of peer cardholders; generating a personalized lending score associated with the candidate cardholder based at least in part on the normalized transaction data, wherein the personalized lending score indicates a spending trend of the candidate cardholder; and transmitting the personalized lending score to the requestor computing device.

11 . The method of claim 10 , further comprising:
applying a plurality of predetermined ratings to the spending categories of the transaction data, wherein a negative rating indicates a spending category associated with an undesirable spending habit and a positive rating indicates a spending category associated with a desirable spending habit; and generating the personalized lending score based at least in part on the spending category of the transaction data and the predetermined ratings.

12 . The method of claim 11 , wherein a net rating for all spending categories is equal to 1.

13 . The method of claim 10 , wherein determining the demographic data includes determining an age group of the cardholder, an income group of the cardholder, a geographical residence location of the cardholder, and combinations thereof.

14 . The method of claim 10 , wherein generating the personalized lending score includes generating an expense rating wherein a negative expense rating indicates that the candidate cardholder's spending is dominated by undesirable spending categories and a positive expense rating indicates that the candidate cardholder's spending is dominated by desirable spending categories.

15 . The method of claim 10 , wherein generating the personalized lending score includes generating a total expense rating wherein a negative total expense rating indicates that the candidate cardholder generally spends more than a peer cardholder associated with the demographic data, and a positive total expense rating indicates that the candidate cardholder generally saves more than a peer cardholder associated with the demographic data.

16 . The method of claim 10 , wherein the plurality of spending categories includes one or more of the following spending categories: groceries, high end groceries, low end groceries, fast food restaurants, fine dining restaurants, healthy eating restaurants, travel, entertainment, gamblin

Assignees

• Mastercard International Inc

Inventors

• Bhattacharjee Manash
• Tiwaree Prashanna S
• Sharma Prashant

Classifications

• G06Q40/03 Primary

  Credit; Loans; Processing thereof · CPC title

• G06Q40/02 Primary

  Banking, e.g. interest calculation or account maintenance (credit or loans G06Q40/03) · CPC title

• G06Q30/0204

  Market segmentation · CPC title

• G06Q40/025 Primary

  Physics · mapped topic