Score based decisioning

• US 7,251,624 B1
• Filed: 09/29/2000
• Issued: 07/31/2007
• Est. Priority Date: 09/08/1992
• Status: Expired due to Fees

• Current Assignee: Fair Isaac Corporation
• Original Assignee: Fair Isaac Corporation
• Inventors: Milana, Joseph P., Wilhelm, Wesley K., Lee, Walter W., Shoham, Daniel, Jost, Alan P.
• Primary Examiner(s): Gilligan; C. Luke
• Assistant Examiner(s): Najarian; Lena
• Application Number:
  US09/675,992
• Time in Patent Office: 2,496 Days
• Field of Search: 709/203, 705/10, 705/35, 705/38, 705/26, 706/21, 715/700
• US Class Current: 705/35
• CPC Class Codes: G06Q 30/06 Buying, selling or leasing transactions
• CPC Class Codes: G06Q 30/0601 Electronic shopping [e-shopping]
• CPC Class Codes: G06Q 30/0609 Buyer or seller confidence or verification
• CPC Class Codes: G06Q 40/00 Finance; Insurance; Tax strategies; Processing of corporate or income taxes
• CPC Class Codes: G06Q 40/03 Credit; Loans; Processing thereof
• CPC Class Codes: G06Q 40/08 Insurance

Abstract

The invention aids an entity to selectively request additional data about a user to serve the needs of the entity without causing the user to have an adverse reaction. The invention helps an entity to determine the condition under which to request additional data so as to reduce the likelihood of causing the user to have an adverse reaction, e.g., terminate the interaction. One embodiment of the invention concerns customers requesting transactions with on-line merchants. More specifically, this embodiment aids merchants by detecting Internet credit card transactions that are likely to be fraudulent, and providing the merchants with mechanisms for managing a suspected transaction as it occurs to obtain additional information that can be useful to reducing the likelihood of fraud.

Claims

1. A computer-implemented method for processing a request for a transaction over a client-server network, the method comprising:

receiving a request for a transaction from a customer through a client computer, the request including a first set of transaction data for variables descriptive of the initial transaction;

responsive to receiving said request, generating a transaction score using a prediction model based on the first set of transaction data and based on the customer'"'"'s profile data that contain summaries of historical data that include prior customer transaction data, where the transaction score is indicative of a level of risk associated with the transaction; and

responsive to the generated transaction score, performing;

indicating to the client computer that the request for the transaction is rejected if the transaction score is above an upper bound;

proceeding with the request for a transaction if the transaction score is below a lower bound; and

obtaining additional data from the customer if the transaction score is between the lower bound and upper bound inclusive, said obtaining additional data comprising;

determining, for each of a plurality of follow-up question sets, a probability of non-attrited fulfillment of the transaction after presentment of the follow-up question set, said determining the probability based on a metric for the value of additional data, using information provided by the question set, the transaction information, and the customer'"'"'s profile data, and said determining the probability based on a likelihood that the customer will terminate the transaction if the customer is presented with the follow-up question set; and

selecting a follow-up question set, from among the plurality of follow-up question sets, with the greatest probability of non-attrited fulfillment of the transaction, and presenting the selected follow-up question set to the customer.

2. The computer-implemented method of claim 1, wherein at least one of the upper bound and the lower bound is a function of the value of the transaction.

3. The computer-implemented method of claim 1, wherein receiving a request for a transaction comprises:

generating and forwarding to the customer'"'"'s client computer a form for obtaining the first set of transaction data.

4. The computer-implemented method of claim 1, wherein obtaining additional data from the customer comprises:

generating and forwarding a request for additional data to the customer'"'"'s client computer.

5. The computer-implemented method of claim 4, wherein generating and forwarding to the customer'"'"'s client computer a request for additional data comprises generating and forwarding to the customer'"'"'s client computer a form for obtaining additional data.

6. A system for processing a request for a transaction over a computer network, the system comprising:

a transaction-scoring means for receiving transaction data and generating a transaction score using a prediction model based on the received transaction data and based on profile data that contain summaries of historical data that include prior customer transaction data;

a thresholding means for receiving the transaction score and, based on the transaction score, applying the score to at least one threshold to selectively perform;

completing the transaction if the transaction score is below a lower bound;

indicating that the transaction is rejected if the transaction score is above an upper bound;

obtaining additional information from the customer if the transaction score is between the lower bound and upper bound inclusive;

an information value prediction model adapted for receiving data representing a plurality of follow-up question sets and for determining, for each of the plurality of follow-up question sets, a metric for the value of additional information provided by the follow-up question set;

a friction model adapted for receiving data representing the plurality of follow-up question sets and for determining, for each of the plurality of follow-up question sets, the likelihood that a customer will terminate a transaction if presented with the follow-up question set; and

a question set optimization (a) means for determining, for each of the plurality of follow-up question sets, the probability of non-attrited fulfillment of the transaction after presentment of the follow-up question set, said determining the probability based on the metric for the value of additional information provided by the follow-up question set, the transaction, and the customer'"'"'s profile, said determining the probability based on the likelihood that a user customer will terminate the transaction if the user customer is presented with the follow-up question set, and (b) means for selecting the follow-up question set with the highest probability of non-attrited fulfillment of the transaction and means for presenting the selected follow-up question set to the customer.