Identifying Users' Identity Through Tracking Common Activity

• US 20170078415A1
• Filed: 09/15/2015
• Published: 03/16/2017
• Est. Priority Date: 09/15/2015
• Status: Active Grant

• Current Assignee: Yahoo AD Tech LLC
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Xia, Zhen, Wen, Musen, Wang, Liang, Chen, Datong
• Granted Patent: US 10,715,612 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

Systems and methods for identifying users according to their activity are disclosed. The identification of a user includes accessing a user activity log having a plurality of identifiers and corresponding activity information for each identifier, determining identifiers having correlating activity information, and assigning identifiers having correlating activity information to a common user.

Claims

1. A method of identifying a user, comprising:

accessing a user activity log comprising a plurality of identifiers and corresponding activity information for each identifier;

determining identifiers having correlating activity information; and

assigning identifiers having correlating activity information to a common user.

2. The method of claim 1, wherein determining identifiers having correlating activity information comprises:

filtering the user activity log to exclude activity associated with popular activities;

building an inverted index of the filtered user activity log, the inverted index having the activity information as a key and set of identifiers associated with the activity as data;

enumerating the possible combinations of identifiers to form potential user groups;

scoring each of the potential user groups based on correlating activity information

3. The method of claim 1 wherein the identifiers are selected from the group consisting of hardware identifiers, cookies, and device profiles.

4. The method of claim 1, wherein the activity information comprises information selected from the group consisting of internet browsing activity, online shopping activity, and internet app usage.

5. The method of claim 2, wherein enumerating the possible combinations of identifiers comprises:

enumerating a limited number of user combinations;

grouping the user combinations according to common sites; and

merging user combinations having common sites.

6. The method of claim 2, wherein filtering the activity data comprises ranking the activity information according to inverse popularity and selecting only those internet activities having a ranking higher than a threshold.

7. The method of claim 2, wherein the user groups are scored according to a quantity of common websites.

8. The method of claim 2, wherein the user groups are scored according to a sum of the inverse document frequency for each common website.

9. A system for identifying a group of activities having a common user, the system comprising:

a processor configured to implement computer instructions;

a first data store storing data representing correlations of identifiers and activity associated with the identifiers; and

a second data store storing computer executable instructions, that when implemented by the processor, cause the system to perform functions comprising;

accessing first data store to retrieve the data representing correlations of identifiers and activity associated with the identifiers;

determining identifiers having correlating activity; and

assigning identifiers having correlating activity to a common user.

10. The system of claim 9, wherein determining identifiers having correlating activity comprises:

filtering the data to exclude activity associated with popular activities;

building an inverted index of the filtered data, the inverted index having the activity as a key and set of identifiers associated with the activity as data;

enumerating the possible combinations of identifiers to form potential user groups;

scoring each of the potential user groups based on correlating activity information.

11. The system of claim 9 wherein the identifiers are selected from the group consisting of hardware identifiers, cookies, and device profiles.

12. The system of claim 9, wherein the activity information comprises information selected from the group consisting of internet browsing activity, online shopping activity, and internet app usage.

13. The system of claim 10, wherein enumerating the possible combinations of identifiers comprises:

enumerating a limited number of user combinations;

grouping the user combinations according to common sites; and

merging user combinations having common sites.

14. The system of claim 10, wherein filtering the activity data comprises ranking the activity information according to inverse popularity and selecting only those internet activities having a ranking higher than a threshold.

15. The system of claim 10, wherein the user groups are scored according to a quantity of common websites.

16. The system of claim 10, wherein the user groups are scored according to a sum of the inverse document frequency for each common website.

17. A computer readable storage media storing data comprising:

a plurality of identifiers; and

data linking groups of the plurality of identifiers based on the historical activity of the identifiers.

18. The computer readable storage media of claim 17 wherein the plurality of identifiers comprise cookies.

19. The computer readable storage media of claim 17, wherein the activity of the identifiers comprises browsing history.