Expert credit recommendation method and system

• US 5,262,941 A
• Filed: 03/30/1990
• Issued: 11/16/1993
• Est. Priority Date: 03/30/1990
• Status: Expired due to Term

• Current Assignee: Deutsche Bank Americas Holding Corporation (Deutsche Bank)
• Original Assignee: ITT Corporation (ITT, Inc.)
• Inventors: Saladin, Emery F., Ruhlin, Kurt A., Gers, Harvey, Mate, Karol V.
• Primary Examiner(s): Envall, Jr., Roy N.
• Assistant Examiner(s): BAI, ARI
• Application Number:
  US07/502,850
• Time in Patent Office: 1,327 Days
• Field of Search: 364/401, 364/406, 364/408
• 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 expert floorplan credit recommendation method and system, comprising a database and stored program for arriving at a credit recommendation by processing data on a computer by means of a decision matrix tree that emulates the thought processes of credit experts.

Claims

1. A computerized expert credit recommendation system useful for credit analysis automation and training credit analysts, comprising:

a) an input means for entering data in response to screen messages;

b) an output means;

c) a programmable digital computer in association with

(i) a database management computer program, which prompts in a predetermined sequence of queries which emulate the reasoning of an expert credit analyst through the output means, a user to enter through the input means, in a predetermined order, data relating to criteria used in making credit recommendations, and which manages such input data;

(ii) a knowledge base containing a plurality of predetermined and defined standardized rating values relating to data on criteria

(iii) an expert credit recommendation computer program;

which manipulates and performs operations on the input data in association with the database management program and the knowledge base

wherein the digital computer utilizes the expert credit recommendation computer program in conjunction with the data input by the user to assign to each of the criteria a standardized rating value from the knowledge base and to utilize such assigned standardized rating value to simulate a plurality of decision matrices arranged in a defined decision matrix tree series, having a financial branch and a dealer rating branch;

wherein, for each branch point in the matrix tree series the system determines and assigns a defined standardized rating value to a resulting criterion;

which standardized rating value is applied in a subsequent decision matrix until a final decision matrix is reached; and

wherein after all the data has been entered by the user in the predetermined sequence of queries, the system generates an overall rating which is displayed to the user on the output means specified by the standardized rating value assigned to a final criterion resulting from the financial branch and the standardized rating value assigned to a final criterion resulting from the dealer rating branch.

2. A computerized expert credit recommendation system as in claim 1, wherein the system further comprises a means for determining a credit recommendation by comparing the defined standardized rating value assigned to a net hard collateral criterion against the overall rating so as to determine and assign a defined standardized rating value to a final rating criterion, which final rating criterion'"'"'s assigned defined standardize rating value is compared against the standardized rating value assigned to a credit recommendation matrix input criterion to arrive at the credit recommendation;

and a means for communicating the credit recommendation to the output means.

3. A computerized expert credit recommendation system as in claim 2, wherein the system further comprises a means for formulating and communicating to the output means comments on the credit recommendation.

6. A method of determining a credit recommendation as in claim 9, wherein the overall rating is compared against the standardized rating value assigned to a net hard collateral criterion so as to determine a standardized rating value for a final rating criterion;

and

comparing the standardized rating value assigned to the final rating criterion against the standardized rating value assigned to a credit recommendation matrix input criterion so as to produce the credit recommendation.

7. A method of determining a credit recommendation as in claim 6, further comprising communicating the credit recommendation to the output means.

9. A computerized expert credit recommendation system as in claim 1, wherein the system operates to simulate decision matrices configured in a defined matrix tree series as in FIGS. 1-4.

4. A method of determining a credit recommendation, comprising:

a) providing to a computer comprising an input means and an output means, a database management program and an expert credit recommendation computer program;

the computer and the expert recommendation computer program operating to simulate a plurality of decision matrices utilizing standardized rating values, said matrices being configured in a defined decision matrix tree series, comprising a financial branch and a dealer rating branch;

wherein the computer and the database management program prompt a user through the output means, via screen prompts generating in a pre-determined sequence and communicated to the output means, to enter data related to specific criteria in a sequence and manner which emulates the reasoning of an expert credit analyst;

b) entering the data related to the criteria through the input means, in response to the prompts;

c) determining and assigning a defined standardized rating value to the data relating to the criteria for which data has been entered, and applying the standardized rating values assigned to those criteria and to the resulting criteria to the decision matrices in a predetermined manner according to the defined decision matrix tree series until a final decision matrix is reached; and

d) comparing the defined standardized rating value assigned to a final resulting criterion from the financial branch against a final resulting criterion from the dealer rating branch in the final decision matrix to arrive at an overall rating.

5. A method of determining a credit recommendation as in claim 4, further comprising communicating the overall rating to the output means.

8. A method of determining a credit recommendation as in claim 4, wherein the financial branch comprises an interim balance sheet decision matrix, an interim income trend decision matrix, and interim income statement decision matrix, an interim balance sheet and trend decision matrix, an interim income statement and trend decision matrix, an interim financial decision matrix, a balance sheet decision matrix, an income trend decision matrix, an income statement decision matrix, a balance sheet and trend decision matrix, an income statement and trend decision matrix, and a fiscal-year-end decision matrix;

and

the dealer rating branch comprises a lender experience decision matrix, an owner experience decision matrix, a business pay habits decision matrix, a business experience decision matrix, a business character decision matrix, a character decision matrix, a manufacturing strength decision matrix, a dealer collateral decision matrix, a manufacturing collateral decision matrix and a collateral decision matrix.