SYSTEM AND METHOD FOR EXTRAPOLATING DATA FROM A SAMPLE SET

• US 20080256018A1
• Filed: 04/12/2007
• Published: 10/16/2008
• Est. Priority Date: 04/12/2007
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Seth, Amitabh, Chaudhury, Supratim Roy, Hogan, Christopher L.
• Granted Patent: US 8,150,849 B2
• Time in Patent Office: Days
• US Class Current: 707/1
• CPC Class Codes: G06F 16/334 Query execution G06F16/335 takes precedence
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/358 Browsing; Visualisation therefor
• CPC Class Codes: G06Q 30/0241 Advertisements

Abstract

Embodiments of the present invention are directed to identifying a source of bias in a sample set and using statistical techniques to overcome the bias, thereby producing extrapolated data that accurately represents a population including the sample set. The method according to one embodiment comprises identifying a plurality of attributes corresponding to an entity, generating a plurality of clusters from the plurality of attributes, the plurality of clusters including varying numbers of the plurality of attributes, and performing measurements on each of the plurality of clusters to generate measurement data values for each of the plurality of clusters. A cluster with a greatest measurement data value is selected.

Claims

1. A method, comprising:

identifying a plurality of attributes corresponding to an entity;

generating a plurality of clusters from the plurality of attributes, the plurality of clusters including varying numbers of the plurality of attributes;

performing measurements on each of the plurality of clusters to generate measurement data values for each of the plurality of clusters; and

selecting a cluster with a greatest measurement data value.

2. The method according to claim 1, comprising receiving a data set from a data source.

3. The method according to claim 2, comprising:

identifying attributes included in the selected cluster; and

sorting the data set based on the attributes.

4. The method according to claim 2, wherein the data set comprises counter data corresponding to a number of conversions.

5. The method according to claim 3, wherein the sorting comprises generating strata within the data set as a function of the attributes.

6. The method according to claim 4, wherein each stratum within the strata corresponds to one of the attributes.

7. The method according to claim 4, comprising computing a metric for each of the stratum to generate stratum data.

8. The method according to claim 7, wherein the metric is a conversion rate.

9. The method according to claim 7, comprising applying a weight to each of the stratum data to generate weighted stratum data, the weight being a percentage of a total data set represented by the data set.

10. The method according to claim 9, comprising:

computing a sum of the weighted stratum data; and

computing the metric for the total data set using the sum.

11. The method according to claim 10, comprising:

computing the metric for the data set using the sum; and

comparing the metric for the data set to the metric for the total data set to determine a quality of the data set.