Execution of multiple models using data segmentation

• US 6,542,894 B1
• Filed: 12/09/1998
• Issued: 04/01/2003
• Est. Priority Date: 12/09/1998
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: Unica Corporation (International Business Machines Corporation)
• Inventors: Lee, Yuchun, Crites, Robert, Kennedy, Ruby
• Primary Examiner(s): Mizrahi, Diane D.
• Assistant Examiner(s): PARDO, THUY N
• Application Number:
  US09/208,037
• Time in Patent Office: 1,574 Days
• Field of Search: 706/12, 705/7, 705/10, 707/2, 707/5, 707/6, 707/8, 707/3, 707/101, 707/102
• US Class Current: 1/1
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases
• CPC Class Codes: G06F 2216/03 Data mining
• CPC Class Codes: Y10S 707/99932 Access augmentation or optimizing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A method executed on a computer for modeling expected behavior is described. The method includes scoring records of a dataset that is segmented into a plurality of data segments using a plurality of models and converting scores of the records into probability estimates. Two of the techniques described for converting scores into probability estimates are a technique that transforms scores into the probabilities estimates based on an equation and a binning technique that establishes a plurality of bins and maps records based on a score for the record to one of the plurality of bins.

Claims

1. A method of combining lift charts that represent predicted response rates comprises:

establishing a plurality of bins for each lift chart;

assigning records for each lift chart into one of the bins established for its corresponding lift chart;

sorting bins based on an average response rate for each bin; and

combining results of sorting the bins into a summary lift chart representation.

2. The method of claim 1 wherein sorting combines the bins having the highest response rate to the lowest response rate.

3. The method of claim 1 wherein the lift charts represent predicted response rates for a corresponding number of mailing lists.

4. The method of claim 1 wherein data from the lift charts are derived from a corresponding gains table.