AUDIENCE SEGMENT ESTIMATION

• US 20110055009A1
• Filed: 08/28/2009
• Published: 03/03/2011
• Est. Priority Date: 08/28/2009
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Kiveris, Raimondas
• Granted Patent: US 8,370,202 B2
• Time in Patent Office: Days
• US Class Current: 705/14.52
• CPC Class Codes: G06F 16/23 Updating
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0246 Traffic
• CPC Class Codes: G06Q 30/0255 based on user history

Abstract

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for estimating segments from key values for advertising impressions.

Claims

1. A computer-implemented method, comprising:

accessing, by a data processing apparatus, publisher logs including data describing past impressions delivered for pageviews for publisher sites, the data including, for each impression;

a user identifier;

a publisher identifier identifying a publisher;

a pageview value identifying a pageview; and

a plurality of publisher key values of the publisher provided for the impression;

generating, by the data processing apparatus, from the publisher logs publisher pageview data for each publisher, the publisher pageview data for each publisher comprising pageview records representative of pageviews that occurred during a first time period, each pageview record representing one or more pageviews associated with a user identifier and a publisher identifier pair during a second time period that occurred during the first time period, each pageview record including;

a user identifier;

a publisher identifier; and

publisher key values provided for the impressions that occurred for the one or more pageviews represented by the pageview record;

receiving, at the data processing apparatus, a segment definition for a publisher, the segment definition defining a segment by one or more publisher key values of the publisher;

assigning, by the data processing apparatus, to the segment user identifiers from records that include publisher key values that satisfy the segment defined by the segment definition;

determining, by the data processing apparatus, one or more segment characteristics based on the user identifiers assigned to the segment and the pageview records of the user identifiers assigned to the segment; and

providing, in response to the segment definition, response data describing the one or more characteristics of the segment.

2. The computer-implemented method of claim 1, wherein each pageview record includes, for each pageview the pageview record represents, a pageview/key value map that indicates which of the publisher key values were provided for impressions for the pageview.

3. The computer-implemented method of claim 2, wherein determining one or more segment characteristics comprises determining a first number of pageviews from the pageview records of the user identifiers assigned to the segment.

4. The computer-implemented of claim 3, wherein:

determining the first number of pageviews from the pageview records of the user identifiers assigned to the segment comprise determining the number pageview/key value maps that indicate publisher key values that satisfy the segment definition; and

further comprising;

determining a second number of pageviews from the pageview records of the user identifiers assigned to the segment, the second number of pageviews equal to the total number of pageviews represented by the pageview records of the user identifiers assigned to the segment; and

determining a segment quality measure that is proportional to a ratio of the second number of pageviews to the first number of pageviews.

5. The computer-implemented method of claim 4, further comprising:

generating segment population data, the segment population data being an association of user identifiers assigned to the segment defined by the segment definition data and data defining the segment; and

providing, in response to the segment definition, the segment population data to a publisher to facilitate advertisement targeting for future pageviews associated with the user identifiers assigned to the segment.

6. The computer-implemented method of claim 2, wherein:

for each pageview record, the publisher key values are stored in respective fields in the pageview record, each respective field associated with a respective index value;

the pageview/key value map for each of the one or more pageviews represented by the pageview record is a pageview/key value bit map having a corresponding first value in an indexed position corresponding to a respective index value of a publisher key value that was provided for an impression for the pageview, and having a corresponding second value in an indexed position corresponding to a respective index value of a publisher key value that was not provided for the impression for the pageview.

7. The computer-implemented method of claim 6, wherein assigning to the segment user identifiers from records that include publisher key values that satisfy the segment defined by the segment definition comprises:

generating from the segment definition a bit mask for each pageview record, the bit mask having the corresponding first value in an indexed position corresponding to a respective index value of a publisher key value in the pageview record, and the corresponding second value in an indexed position corresponding to the respective index value of a publisher key value not in the pageview record; and

for each pageview record for the publisher, comparing the segment definition bit mask of the pageview record to each of the pageview/key value bit maps of the pageview record.

8. The computer-implemented method of claim 1, wherein:

generating segment population data, the segment population data being an association of user identifiers assigned to the segment defined by the segment definition data and data defining the segment; and

providing, in response to the segment definition, the segment population data to a publisher to facilitate advertisement targeting for future pageviews associated with the user identifiers assigned to the segment.

9. A system, comprising:

a data storage apparatus storing publisher logs including data describing past impressions delivered for pageviews of publisher sites, the data including, for each impression;

a user identifier;

a publisher identifier identifying a publisher;

a pageview value identifying a pageview; and

a plurality of publisher key values of the publisher provided for the impression;

instructions stored in a computer memory device and comprising instructions executable by a data processing apparatus and upon execution cause the data processing apparatus to perform operations comprising;

generating from the publisher logs publisher pageview data for each publisher and store the pageview data in the data storage apparatus, the publisher pageview data for each publisher comprising pageview records representative of pageviews that occurred during a first time period, each pageview record representing one or more pageviews associated with a user identifier and a publisher identifier pair during a second time period that occurred during the first time period, each pageview record including;

a user identifier;

a publisher identifier; and

publisher key values provided for the impressions that occurred for the one or more pageviews represented by the pageview record;

receiving a segment definition for a publisher, the segment definition defining a segment by one or more publisher key values of the publisher;

assigning to the segment user identifiers from records that include publisher key values that satisfy the segment defined by the segment definition;

determining one or more segment characteristics based on the user identifiers assigned to the segment and the pageview records of the user identifiers assigned to the segment; and

providing, in response to the segment definition, response data describing the one or more characteristics of the segment.

10. The system of claim 9, wherein each pageview record includes, for each pageview the pageview record represents, a pageview/key value map that indicates which of the publisher key values were provided for impressions for the pageview.

11. The system of claim 10, wherein determining one or more segment characteristics comprises determining a first number of pageviews from the pageview records of the user identifiers assigned to the segment.

12. The system of claim 11, wherein:

determining the first number of pageviews from the pageview records of the user identifiers assigned to the segment comprise determining the number pageview/key value maps that indicate publisher key values that satisfy the segment definition; and

wherein the instructions cause the data processing apparatus to perform further operations comprising;

determining a second number of pageviews from the pageview records of the user identifiers assigned to the segment, the second number of pageviews equal to the total number of pageviews represented by the pageview records of the user identifiers assigned to the segment; and

determining a segment quality measure that is proportional to a ratio of the second number of pageviews to the first number of pageviews.

13. The system of claim 12, wherein:

generating segment population data, the segment population data being an association of user identifiers assigned to the segment defined by the segment definition data and data defining the segment; and

providing, in response to the segment definition, the segment population data to a publisher to facilitate advertisement targeting for future pageviews associated with the user identifiers assigned to the segment.

14. The system of claim 10, wherein:

for each pageview record, the publisher key values are stored in respective fields in the pageview record, each respective field associated with a respective index value;

the pageview/key value map for each of the one or more pageviews represented by the pageview record is a pageview/key value bit mask having a corresponding first value in an indexed position corresponding to a respective index value of a publisher key value that was provided for an impression for the pageview, and having a corresponding second value in an indexed position corresponding to a respective index value of a publisher key value that was not provided for the impression for the pageview.

15. The system of claim 14, wherein assigning to the segment user identifiers from records that include publisher key values that satisfy the segment defined by the segment definition comprises:

generating from the segment definition a bit mask for each pageview record, the bit mask having the corresponding first value in an indexed position corresponding to a respective index value of a publisher key value in the pageview record, and the corresponding second value in an indexed position corresponding to the respective index value of a publisher key value not in the pageview record; and

for each pageview record for the publisher, comparing the segment definition bit mask of the pageview record to each of the pageview/key value bit maps of the pageview record.

16. A system, comprising:

a data storage apparatus storing publisher pageview data for a plurality of publishers, the publisher pageview data for each publisher comprising pageview records representative of pageviews that occurred during a first time period, each pageview record representing one or more pageviews associated with a user identifier and a publisher identifier pair, each pageview record including;

a user identifier;

a publisher identifier identifying a publisher; and

publisher key values provided for the impressions that occurred for the one or more pageviews represented by the pageview record;

wherein for each pageview record, the publisher key values are stored in respective fields in the pageview record, each respective field associated with a respective index value, and each pageview record includes, for each pageview the pageview record represents, a pageview/key value map that indicates which of the publisher key values were provided for impressions for the pageview, the pageview/key value map for each of the one or more pageviews represented by the pageview record is a pageview/key value bit map having a corresponding first value in an indexed position corresponding to a respective index value of publisher key value that was provided for an impression for the pageview, and having a corresponding second value in an indexed position corresponding to a respective index value of a publisher key value that was not provided for the impression for the pageview.

instructions stored in a computer memory device and comprising instructions executable by a data processing apparatus and upon execution cause the data processing apparatus to perform operations comprising;

receiving a segment definition for a publisher, the segment definition defining a segment by one or more publisher key values of the publisher;

generating from the segment definition a bit mask for each pageview record, the bit mask having the corresponding first value in an indexed position corresponding to a respective index value of a publisher key value in the segment definition, and the corresponding second value in an indexed position corresponding to the respective index value of a publisher key value not in the pageview record;

for each pageview record for the publisher, comparing the segment definition bit mask of the pageview record to each of the pageview/key value bit maps of the pageview record to determine which pageviews satisfy the segment definition;

determining one or more segment characteristics based on the pageviews that are determined to satisfy the segment definition; and

providing, in response to the segment definition, response data describing the one or more characteristics of the segment.

17. A method, comprising:

generating, by a data processing apparatus, publisher pageview data for a plurality of publishers, the publisher pageview data for each publisher comprising pageview records representative of pageviews that occurred during a first time period, each pageview record representing one or more pageviews associated with a user identifier and a publisher identifier pair, each pageview record including;

a user identifier;

a publisher identifier identifying a publisher; and

publisher key values provided for the impressions that occurred for the one or more pageviews represented by the pageview record;

receiving, at the data processing apparatus, a segment definition for a publisher, the segment definition defining a segment by one or more publisher key values of the publisher;

determining, by the data processing apparatus, pageviews for which publisher key values that satisfy the segment defined by the segment definition were provided for impressions;

determining, by the data processing apparatus, one or more segment characteristics based on the pageviews that are determined to satisfy the segment definition; and

providing, in response to the segment definition, response data describing the one or more characteristics of the segment.

18. A system, comprising:

means for publisher pageview data for a plurality of publishers, the publisher pageview data for each publisher comprising pageview records representative of pageviews that occurred during a first time period, each pageview record representing one or more pageviews associated with a user identifier and a publisher identifier pair; and

means for determining pageviews for which publisher key values that satisfy the segment defined by the segment definition were provided for impressions and for determining one or more segment characteristics based on the pageviews that are determined to satisfy the segment definition.

19. A system, comprising:

means for publisher pageview data for a plurality of publishers, the publisher pageview data for each publisher comprising pageview records representative of pageviews that occurred during a first time period, each pageview record representing one or more pageviews associated with a user identifier and a publisher identifier pair; and

means for assigning to the segment user identifiers from records that include publisher key values that satisfy the segment defined by the segment definition and for determining one or more segment characteristics based on the user identifiers assigned to the segment and the pageview records of the user identifiers assigned to the segment.