Automated method and system for predicting and/or verifying income

• US 20070055621A1
• Filed: 09/01/2006
• Published: 03/08/2007
• Est. Priority Date: 09/01/2005
• Status: Abandoned Application

• Current Assignee: First Advantages Corporation
• Original Assignee: First Advantage Corporation (Symphony Technology Group LLC)
• Inventors: Gothe, Per, Tischler, Howard, Riordan, Chip
• Application Number:
  US11/514,171
• Publication Number:
  US 20070055621A1
• Time in Patent Office: Days
• US Class Current: 705/38
• CPC Class Codes: G06Q 40/02 Banking, e.g. interest calculation or account maintenance credit or loans G06Q40/03
• CPC Class Codes: G06Q 40/03 Credit; Loans; Processing thereof

Abstract

An income predictor deploys an income model that is based on credit information obtained for a representative group of the general population. The credit information is matched against actual employment and income information to predict an income range having a specified confidence factor. The income predictor can be used by subscribers (e.g., creditors, employers, etc.) to predict and/or verify the income of a candidate (e.g., applicants, potential employees, etc.).

Claims

1. A method of predicting income, comprising:

creating an income model by evaluating a number of records that contain actual income information, wherein the number of records is a representative sample of an actual population;

accessing a credit report for a candidate; and

applying the income model to the credit report to generate a prediction of an income of the candidate, wherein the prediction is statistically significant.

2. The method of claim 1, further comprising:

sending the prediction to a third party for verification.

3. The method of claim 2, further comprising:

refining the income model if the prediction is not accurate based on the verification by the third party.

4. A method of approving a loan, comprising:

predicting an income of a candidate, wherein the predicting comprises;

(a) creating an income model by evaluating a number of records that contain actual income information, wherein the number of records is representative of an actual population;

(b) accessing a credit report for the candidate; and

(c) applying the income model to the credit report to generate a prediction of the income of the candidate, wherein said prediction is statistically significant;

receiving an income statement from the candidate;

comparing the income statement with the prediction to determine whether a difference between the income statement and the prediction is within a predefined tolerance; and

verifying the income statement or the prediction via a third party if the difference is greater than the predefined tolerance.

5. The method of claim 4, further comprising:

taking no further action if the difference is within the predefined tolerance.

6. A computer program product comprising a computer useable medium having computer readable program code functions embedded in said medium for causing a computer to predict income, comprising:

a first computer readable program code function that causes the computer to create an income model by evaluating a number of records that contain actual income information, wherein the number of records is representative of an actual population;

a second computer readable program code function that causes the computer to access a credit report for a candidate; and

a third computer readable program code function that causes the computer to apply the income model to the credit report to generate a prediction of an income of the candidate, wherein the prediction is statistically significant.

7. The computer program product of claim 6, further comprising:

a fourth computer readable program code function that causes the computer to send the prediction to a third party for verification.