Dynamic normalization of internet traffic

• US 9,900,395 B2
• Filed: 01/16/2015
• Issued: 02/20/2018
• Est. Priority Date: 01/27/2012
• Status: Active Grant

• Current Assignee: comScore, Inc.
• Original Assignee: comScore, Inc.
• Inventors: James, Derrick, Novick, Jonathan H.
• Primary Examiner(s): Meky, Moustafa M
• Application Number:
  US14/598,363
• Publication Number:
  US 20150201031A1
• Time in Patent Office: 1,131 Days
• CPC Class Codes: G06Q 10/04 Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" market predictions or forecasting for commercial activities G06Q30/0202
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: H04L 43/0876 Network utilisation, e.g. volume of load or congestion level
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

The disclosure herein concerns a method of estimating Internet traffic including taking a size of a target population based on at least one of census data and third party data, identifying a percentage of the target population that displays an online behavior, receiving clickstream data at a host processing facility from a clickstream panel that connotes Internet activity of individual members of the clickstream panel, identifying a fraction of participants within the clickstream panel that exhibit the online behavior, and producing, at the host processing facility, an estimate of the target population'"'"'s Internet activity by first scaling the clickstream data for the participants in the clickstream panel exhibiting the online behavior so that it matches the percentage in the target population and then scaling the data for all members of the clickstream panel by the relative size of the target population.

Claims

1. A method of estimating Internet traffic, the method comprising:

determining a size of a target population based on a data source;

identifying a fraction of participants within a clickstream panel that is managed by a party other than a publisher of the data source that were active in a predetermined time period;

receiving clickstream data at a host from the party that connotes Internet activity of the fraction of participants within the clickstream panel; and

determining, at the host, an estimate of the target population'"'"'s Internet activity by scaling the clickstream data for each member of the clickstream panel by a scaling factor that is based on the size of the target population relative to the fraction of the participants within the clickstream panel active during the predetermined time period.

2. The method of claim 1, further comprising refining the estimate of the target population'"'"'s Internet activity by applying a bias adjustment to the estimate.

3. The method of claim 2, wherein the bias adjustment includes a structural bias adjustment.

4. The method of claim 3, wherein the structural bias adjustment is based on a daily activity that is overrepresented and underrepresented due to a makeup of a sample population.

5. The method of claim 2, wherein applying the bias adjustment to the estimate comprises adjusting the fraction of participants active during the predetermined time period based on the predetermined time period being at least one of a holiday, a day of a week, a day in an identified season, and a time period that is subject to an identified exogenous effect.

6. The method of claim 5, wherein the identified exogenous effect is at least one of a weather event, a news event, a sporting event, an act of nature, and an act of war.

7. The method of claim 2, wherein the bias adjustment is calculated using generalized linear regression models.

8. The method of claim 1, wherein the Internet activity comprises clicking, dwelling, or viewing of individual members of the clickstream panel.

9. The method of claim 1, wherein the data source is at least one of census data and third party data.

10. A method of estimating Internet traffic, the method comprising:

determining a size of a target population based on a data source;

taking a number of devices in use by each member of the target population to access the Internet;

receiving clickstream data at a host from a party other than a publisher of the data source that connotes Internet activity of members of a clickstream panel;

calculating a device weighting factor for a clickstream by dividing a number of devices expected to be in use by each member of the target population by a number of devices that contributed to a user'"'"'s clickstream; and

determining, at the host, an estimate of the target population'"'"'s Internet activity by scaling the clickstream data for each member of the clickstream panel by a scaling factor that is based on the size of the target population and the device weighting factor.

11. The method of claim 10, wherein the number of devices in use by each member of the target population to access the Internet is determined through a survey.

12. The method of claim 10, wherein the Internet activity comprises clicking, dwelling, or viewing of individual members of the clickstream panel.

13. The method of claim 10, wherein the data source is at least one of census data and third party data.

14. A method of estimating Internet traffic, the method comprising:

determining a size of a target population based on a data source;

identifying a percentage of the target population that displays an online behavior;

receiving clickstream data at a host from a party other than a publisher of the data source that connotes Internet activity of all members of a clickstream panel;

identifying a fraction of members of the clickstream panel that exhibit the online behavior;

calculating a scaling factor for the fraction of members of the clickstream panel by dividing the percentage of the target population that displays the online behavior by the fraction of members of the clickstream panel that exhibit the online behavior; and

determining, at the host, an estimate of the target population'"'"'s Internet activity by scaling clickstream data for members of the clickstream panel exhibiting the online behavior by the scaling factor and scaling the clickstream data for all members of the clickstream panel by a relative size of the target population.

15. The method of claim 14, wherein the online behavior includes at least one of a website visited, a URL visited, a category of websites/URLs visited, an activity engaged in online, a shopping behavior, an entertainment behavior, a news behavior, a reading behavior, a financial services behavior, an interaction with a bank or financial service provider, watching a show or movie, using a specific email provider, streaming music, conducting a search query, and an activity during a predetermined time period.

16. The method of claim 14, wherein the data source is at least one of census data and third party data.

17. The method of claim 14, wherein the Internet activity comprises clicking, dwelling, or viewing of individual members of the clickstream panel.

18. The method of claim 14, wherein the data source is at least one of census data and third party data.