Data fusion using behavioral factors

• US 8,838,601 B2
• Filed: 08/31/2011
• Issued: 09/16/2014
• Est. Priority Date: 08/31/2011
• Status: Active Grant

• Current Assignee: comScore, Inc.
• Original Assignee: comScore, Inc.
• Inventors: Meierhoefer, Cameron S., Kapar, David Pham, Kemp, Patrick James, Vaughn, Alan
• Primary Examiner(s): Beausoliel, Jr., Robert
• Assistant Examiner(s): Liu, Hexing
• Application Number:
  US13/222,647
• Publication Number:
  US 20130054628A1
• Time in Patent Office: 1,112 Days
• Field of Search: None
• US Class Current: 707/737
• CPC Class Codes: G06F 16/00 Information retrieval; Database structures therefor; File system structures therefor
• CPC Class Codes: G06F 16/284 Relational databases
• CPC Class Codes: G06Q 30/0203 Market surveys; Market polls
• CPC Class Codes: G06Q 30/0241 Advertisements
• CPC Class Codes: H04L 67/10 in which an application is distributed across nodes in the network software deployment G06F8/60; multiprogramming arrangements G06F9/46

Abstract

A first data set associated with a first group of users is accessed. The first data set includes demographic data, online behavior data, and additional user data associated with the users in the first group. A second data set associated with a second group of users is accessed. The second data set includes demographic data and online behavior data but not additional user data associated with the users in the second group. One or more sets of matched users are determined based on the demographic data and online behavior data included in the first data set and the demographic data and online behavior data included in the second data set. Each set includes a user from the first group matched with a user from the second group. Based on the one or more sets of matched users, an augmented second data set that includes additional user data associated with the users in the second group is generated. One or more reports are generated based on the augmented second data set.

Claims

1. A system comprising:

one or more processing devices; and

one or more storage devices storing instructions that, when executed by the one or more processing devices, cause the one or more processing devices to perform the following operations;

access a first data set associated with a first group of users, the first data set including demographic data, online behavior data, and additional user data associated with the users in the first group;

access a second data set associated with a second group of users, the second data set including demographic data and online behavior data but not additional user data associated with the users in the second group;

group users in the first group and users in the second group into one or more subsets based on the demographic data included in the first data set and the demographic data included in the second data set;

create a vector of online behavior data for each user from the first group grouped in a first subset and for each user from the second group grouped into the first subset;

match a first user in the first group with a second user in the second group by;

determining a distance between the vector created for the second user and each of the vectors created for the users from the first group grouped in the first subset,

determining that the distance between the vector created for the first user and the vector created for the second user is smaller than the distances between the vector created for the second user and the other users from the first group grouped in the subset, and

matching the first user with the second user in response to determining that the distance between the vector created for the first user and the vector created for the second user is smaller than the distances between the vector created for the second user and the other users from the first group grouped in the subset;

generate, based on matching the first user in the first group with the second user in the second group, an augmented second data set that includes additional user data associated with the second user in the second group; and

generate one or more reports based on the augmented second data set.

2. The system of claim 1, wherein the instructions that, when executed by the one or more processing devices, cause the one or more processing devices to generate the one or more reports include instructions that, when executed by the one or more processing devices, cause the one or more processing devices to generate one or more reports based on the first data set and the augmented second data set.

3. The system of claim 1, wherein the instructions that, when executed by the one or more processing devices, cause the one or more processing devices to group users in the first group and users in the second group into one or more subsets include instructions that, when executed by the one or more processing devices, cause the one or more processing devices to:

compare demographic data included in the first data set and the demographic data included in the second data set;

determine, based on the comparison between the demographic data included in the first data set and the demographic data included in the second data set, that demographic data associated with a first user in the first group matches demographic data associated with a second user in the second group; and

group the user in the first group and the user in the second group into a subset.

4. The system of claim 1, wherein the instructions that, when executed by the one or more processing devices, cause the one or more processing devices to determine a distance between the vector created for the second user and each of the vectors created for the users from the first group grouped in the subset include instructions that, when executed by the one or more processing devices, cause the one or more processing devices to determine a Mahalanobis distance between the vector created for the second user and each of the vectors created for the users from the first group grouped in the subset.

5. The system of claim 1, wherein the additional user data included in the first data set includes survey data associated with one or more users in the first group.

6. A computer-implemented method comprising:

accessing a first data set associated with a first group of users, the first data set including demographic data, online behavior data, and additional user data associated with the users in the first group;

accessing a second data set associated with a second group of users, the second data set including demographic data and online behavior data but not additional user data associated with the users in the second group;

grouping users in the first group and users in the second group into one or more subsets based on the demographic data included in the first data set and the demographic data included in the second data set

creating a vector of online behavior data for each user from the first group grouped in a first subset and for each user from the second group grouped into the first subset

matching a first user in the first group with a second user in the second group by;

determining a distance between the vector created for the second user and each of the vectors created for the users from the first group grouped in the first subset,

determining that the distance between the vector created for the first user and the vector created for the second user is smaller than the distances between the vector created for the second user and the other users from the first group grouped in the subset, and

matching the first user with the second user in response to determining that the distance between the vector created for the first user and the vector created for the second user is smaller than the distances between the vector created for the second user and the other users from the first group grouped in the subset;

generating, based on matching the first user in the first group with the second user in the second group, an augmented second data set that includes additional user data associated with the second user in the second group; and

generating one or more reports based on the augmented second data set.

7. The system of claim 6, wherein generating the one or more reports includes generating one or more reports based on the first data set and the augmented second data set.

8. The system of claim 6, wherein grouping users in the first group and users in the second group into one or more subsets includes:

comparing demographic data included in the first data set and the demographic data included in the second data set;

determining, based on the comparison between the demographic data included in the first data set and the demographic data included in the second data set, that demographic data associated with a first user in the first group matches demographic data associated with a second user in the second group; and

grouping the user in the first group and the user in the second group into a subset.

9. The system of claim 6, wherein determining a distance between the vector created for the second user and each of the vectors created for the users from the first group grouped in the subset includes determining a Mahalanobis distance between the vector created for the second user and each of the vectors created for the users from the first group grouped in the subset.

10. The system of claim 6, wherein the additional user data included in the first data set includes survey data associated with one or more users in the first group.

11. A system comprising:

one or more processing devices; and

one or more storage devices storing instructions that, when executed by the one or more processing devices, cause the one or more processing devices to perform the following operations;

access a first set of demographic data for a first group of client systems, the first set of demographic data being determined based on survey data collected from each member of the first group of client systems;

access a second set of demographic data for a second group of client systems, the second set of demographic data being determined based on profile data associated with each member of the second group of client systems;

compare the first set of demographic data with the second set of demographic data;

based on the comparison of the first set of demographic data with the second set of demographic data, group, into subsets, one or more members of the first group of client systems with one or more members of the second group of client systems such that members of each subset share matching demographic data;

access online behavior data for each member of a subset;

compare online behavior data of each member of the subset with online behavior data of other members of the subset by;

creating a vector of online behavior data for each client system from the first group grouped in a first subset and for each client system from the second group grouped into the first subset, the first group grouped in the first subset including a first client system and the second group grouped in the first subset including a second client system,

determining a distance between the vector created for the second client system and each of the vectors created for the client systems from the first group grouped in the first subset, and

determining that the distance between the vector created for the first client system and the vector created for the second client system is smaller than the distances between the vector created for the second client system and the other client systems from the first group grouped in the first subset;

based on the comparison the online behavior data of each member of the subset with online behavior data of the other members of the subset, associate, within the subset, each member of the second group of client systems with a member of the first group of client systems, wherein associating each member of the second group of client systems with a member of the first group of client systems includes matching the first client system with the second client system in response to determining that the distance between the vector created for the first client system and the vector created for the second client system is smaller than the distances between the vector created for the second client system and the other client systems from the first group grouped in the first subset;

for each member of the second group of client systems within the subset, associate survey data collected from the member of the first group of client systems with the member of the second group of client systems with whom the member of the first group of client systems has been associated; and

generate one or more reports based on the association of survey data.

12. The system of claim 11, wherein:

the instructions that, when executed by the one or more processing devices, cause the one or more processing devices to compare the first set of demographic data with the second set of demographic data include instructions that, when executed by the one or more processing devices, cause the one or more processing devices to determine that a first client system from the first group of client systems share one or more of the following demographic variables with a second client system from the second group of client systems;

age, gender, zip code, income, race, ethnicity, whether there are children present in a household associated with the client system, and the size of the household associated with the client system; and

the instructions that, when executed by the one or more processing devices, cause the one or more processing devices to group, into subsets, one or more members of the first group of client systems with one or more members of the second group of client systems such that members of each subset share matching demographic data include instructions that, when executed by the one or more processing devices, cause the one or more processing devices to group the first client system from the first group of client systems and the second client system from the second group of client systems into one of the subsets.

13. The system of claim 11, wherein the instructions that, when executed by the one or more processing devices, cause the one or more processing devices to generate one or more reports based on the association of survey data include instructions that, when executed by the one or more processing devices, cause the one or more processing devices to:

receive one or more target audience variables that are based on the survey data;

search data associated with the first group of client systems and the second group of client systems for the one or more target audience variables; and

generate a report that includes the online behavior data associated with the members of the first group of client systems and the members of the second group of client systems that are associated with survey data that matches the one or more target audience variables.

14. The system of claim 11, wherein the instructions that, when executed by the one or more processing devices, cause the one or more processing devices to associate, within the subset, each member of the second group of client systems with a member of the first group of client systems include instructions that, when executed by the one or more processing devices, cause the one or more processing devices to:

determine the distances between each of the vectors created for the client systems from the second group grouped in the first subset and each of the vectors created for the client systems from the first group grouped in the first subset,

determine that the vector created for a third client system from the first group grouped in the first subset is greater in distance than a vector created for at least one other client system from the first group grouped in the first subset to each of the client systems from the second group grouped in the first subset,

determine a distance between the vector created for the third client system and each of the vectors created for the client systems from the second group grouped in the first subset, and

determine that the distance between the vector created for the third client system and a vector created for a fourth client system from the second group grouped in the first subset is smaller than the distances between the vector created for the third client system and the other client systems from the second group grouped in the first subset; and

the instructions that, when executed by the one or more processing devices, cause the one or more processing devices to associate, within the subset, each member of the second group of client systems with a member of the first group of client systems includes instructions that, when executed by the one or more processing devices, cause the one or more processing devices to match the third client system with the fourth client system in response to determining that the distance between the vector created for the third client system and the vector created for the fourth client system is smaller than the distances between the vector created for the third client system and the other client systems from the second group grouped in the first subset.