Identifying users' identity through tracking common activity

1. A method, comprising:

accessing a user activity log comprising a plurality of identifiers and corresponding activity information for each identifier, the user activity log based on activities of mobile devices and Internet accounts of one or more users;

filtering the user activity log to exclude popular activity information associated with activities exceeding a popularity threshold and generate a filtered user activity log;

building an inverted index of the filtered user activity log, the inverted index having at least some activity information remaining in the user activity log after the filtering as a key and sets of identifiers associated with the at least some activity information as data;

enumerating possible combinations of identifiers to generate a plurality of potential user groups, wherein each potential user group (i) includes a plurality of users, (ii) is associated with a common set of one or more websites, and (iii) is generated based upon a determination that each of the plurality of users of the potential user group are determined to have visited the common set of one or more websites associated with the potential user group, wherein generating the plurality of potential user groups comprises:

generating a first potential user group, including a first plurality of users, in association with a first common set of one or more websites based upon a determination that the first plurality of users visited the first common set of one or more websites, wherein generating the first potential user group comprises grouping a first user of the first plurality of users and a second user of the first plurality of users into the first potential user group based upon (i) a determination that the first user visited a first website of the first common set of one or more websites and a second website of the first common set of one or more websites and (ii) a determination that the second user visited the first website and the second website; and

generating a second potential user group, including a second plurality of users, in association with a second common set of one or more websites based upon a determination that the second plurality of users visited the second common set of one or more websites, wherein generating the second potential user group comprises grouping a third user of the second plurality of users and a fourth user of the second plurality of users into the second potential user group based upon (i) a determination that the third user visited a third website of the second common set of one or more websites and a fourth website of the second common set of one or more websites and (ii) a determination that the fourth user visited the third website and the fourth website;

scoring each potential user group of the plurality of potential user groups based on a quantity of websites in the common set of one or more websites for the potential user group;

determining that users of at least one potential user group of the plurality of potential user groups are associated with a common identity of a user based upon the scoring;

selecting advertisements to be targeted to the user; and

controlling transmission of the advertisements to the user.

2. The method of claim 1 wherein the plurality of identifiers of the user activity log comprise at least one of hardware identifiers, cookies, or device profiles.

3. The method of claim 1 , wherein the corresponding activity information of the user activity log comprises at least one of internet browsing activity, online shopping activity, or internet app usage.

4. The method of claim 1 , 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 two or more user combinations having one or more common sites.

5. The method of claim 1 , wherein filtering the user activity log data comprises ranking at least some second activity information according to inverse popularity and selecting internet activities having a ranking higher than a threshold.

6. The method of claim 1 , wherein scoring each potential user group of the plurality of potential user groups comprises:

scoring the first potential user group with a first score based upon a first quantity of websites in the first common set of one or more websites that the first plurality of users visited; and

scoring the second potential user group with a second score based upon a second quantity of websites in the second common set of one or more websites that the second plurality of users visited.

7. The method of claim 1 , wherein scoring each potential user group of the plurality of potential user groups comprises scoring each potential user group based upon are scored according to a sum of an inverse document frequency for each common website of the common set of one or more websites associated with the potential user group.

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

a processor configured to implement computer instructions; and

memory storing computer executable instructions, that when implemented by the processor, cause the system to perform functions comprising:

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

filtering the data to exclude popular activity information associated with activities exceeding a popularity threshold and generate filtered data;

building an inverted index of the filtered data, the inverted index having at least some activity information remaining in the data after the filtering as a key and sets of identifiers associated with the at least some activity information as data;

enumerating possible combinations of identifiers to generate a plurality of potential user groups, wherein each potential user group (i) includes a plurality of users, (ii) is associated with a common set of one or more websites, and (iii) is generated based upon a determination that each of the plurality of users of the potential user group are determined to have visited the common set of one or more websites associated with the potential user group, wherein generating the plurality of potential user groups comprises:

generating a first potential user group, including a first plurality of users, in association with a first common set of one or more websites based upon a determination that the first plurality of users visited the first common set of one or more websites, wherein generating the first potential user group comprises grouping a first user of the first plurality of users and a second user of the first plurality of users into the first potential user group based upon (i) a determination that the first user visited a first website of the first common set of one or more websites and a second website of the first common set of one or more web sites and (ii) a determination that the second user visited the first website and the second website; and

generating a second potential user group, including a second plurality of users, in association with a second common set of one or more websites based upon a determination that the second plurality of users visited the second common set of one or more websites;

scoring each potential user group of the plurality of potential user groups based on a quantity of websites in the common set of one or more websites for the potential user group;

determining that users of at least one potential user group of the plurality of potential user groups are associated with a common identity of a user based upon the scoring;

selecting content to be targeted to the user; and

controlling transmission of the content to the user.

9. The system of claim 8 wherein the plurality of identifiers of the data comprise at least one of hardware identifiers, cookies, or device profiles.

10. The system of claim 8 , wherein the corresponding activity information of the data comprises at least one of internet browsing activity, online shopping activity, or internet app usage.

11. The system of claim 8 , 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 two or more user combinations having one or more common sites.

12. The system of claim 8 , wherein filtering the activity data comprises ranking at least some second activity information according to inverse popularity and selecting internet activities having a ranking higher than a threshold.

13. The system of claim 8 , wherein scoring each potential user group of the plurality of potential user groups comprises assigning a first score to the first potential user group based upon a determination that the first potential user group is associated with a first quantity of common websites and assigning a second score, greater than the first score, to the second potential user group based upon a determination that the second potential user group is associated with a second quantity of common web sites, wherein the second quantity is greater than the first quantity are scored according to a quantity of common websites.

14. The system of claim 8 , wherein scoring each potential user group of the plurality of potential user groups comprises scoring each potential user group based upon are scored according to a sum of an inverse document frequency for each common website of the common set of one or more websites associated with the potential user group.

15. A non-transitory computer readable storage media comprising computer executable instructions that when executed by a processor perform a method, comprising:

accessing a user activity log comprising a plurality of identifiers and corresponding activity information for each identifier, the user activity log based on activities of mobile devices and Internet accounts of one or more users;

filtering the user activity log to exclude popular activity information associated with activities exceeding a popularity threshold and generate a filtered user activity log;

building an inverted index of the filtered user activity log, the inverted index having at least some activity information remaining in the user activity log after the filtering as a key and sets of identifiers associated with the at least some activity information as data;

enumerating possible combinations of identifiers to generate a plurality of potential user groups, wherein each potential user group (i) includes a plurality of users, (ii) is associated with a common set of one or more websites, and (iii) is generated based upon a determination that each of the plurality of users of the potential user group are determined to have visited the common set of one or more websites associated with the potential user group, wherein generating the plurality of potential user groups comprises:

generating a first potential user group, including a first plurality of users, in association with a first common set of one or more websites based upon a determination that the first plurality of users visited the first common set of one or more websites, wherein generating the first potential user group comprises grouping a first user of the first plurality of users and a second user of the first plurality of users into the first potential user group based upon (i) a determination that the first user visited a first website of the first common set of one or more websites and (ii) a determination that the second user visited the first website; and

generating a second potential user group, including a second plurality of users, in association with a second common set of one or more web sites based upon a determination that the second plurality of users visited the second common set of one or more websites;

scoring each potential user group of the plurality of potential user groups based on a quantity of websites in the common set of one or more websites for the potential user group;

determining that users of at least one potential user group of the plurality of potential user groups are associated with a common identity of a user based upon the scoring;

selecting content to be targeted to the user; and

controlling transmission of the content to the user.

16. The non-transitory computer readable storage media of claim 15 wherein the plurality of identifiers of the user activity log comprise at least one of hardware identifiers, cookies, or device profiles.

17. The non-transitory computer readable storage media of claim 15 , wherein the corresponding activity information of the user activity log comprises at least one of internet browsing activity, online shopping activity, or internet app usage.

18. The non-transitory computer readable storage media of claim 15 , 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 two or more user combinations having one or more common sites.

19. The non-transitory computer readable storage media of claim 15 , wherein filtering the user activity log comprises ranking at least some second activity information according to inverse popularity and selecting internet activities having a ranking higher than a threshold.

20. The non-transitory computer readable storage media of claim 15 , wherein scoring each potential user group of the plurality of potential user groups comprises assigning a first score to the first potential user group associated with a first quantity of common websites and assigning a second score, greater than the first score, to the second potential user group associated with a second quantity of common websites, wherein the second quantity is greater than the first quantity.