Methods and apparatus to estimate demographics of users employing social media

• US 10,333,882 B2
• Filed: 12/27/2013
• Issued: 06/25/2019
• Est. Priority Date: 08/28/2013
• Status: Active Grant

• Current Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Original Assignee: The Nielsen Company LLC (MOOD Media Corporation)
• Inventors: Sheppard, Michael Richard, Reid, Matthew B., Terrazas, Alejandro, Lipa, Peter Otto Robert, Schiller, Brian George
• Primary Examiner(s): Austin, Jamie H
• Application Number:
  US14/142,411
• Publication Number:
  US 20150067075A1
• Time in Patent Office: 2,006 Days
• Field of Search: 705 729, 705 731, 705 732, 705 733
• CPC Class Codes: G06Q 30/00 Commerce
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

Methods, apparatus, systems and articles of manufacture are disclosed to estimate demographics of users employing social media. An example method disclosed herein includes (1) identifying a social media message regarding an asset, the social media message associated with a user identifier associated with the user, (2) determining demographics associated with a group of people exposed to the asset, (3) associating the user identifier with the asset, and (4) repeating (1) to (3). The example method also includes (5) combining demographics associated with two or more different groups of people with which the user identifier is associated to estimate a demographic profile for the user.

Claims

1. A method to estimate demographics of a user of social media, the method comprising:

retrieving, by executing an instruction with a processor of an audience measurement entity, a plurality of electronically transmitted social media messages from a message hosting server based on media referenced in the plurality of electronically transmitted social media messages to reduce a set of social media messages for subsequent processing;

identifying, by executing an instruction with the processor, asset-regarding social media messages of the plurality of electronically transmitted social media messages based on references to assets in respective ones of the social media messages;

identifying, by executing an instruction with the processor, a first social media message of the asset-regarding social media messages mentioning a first asset, the first social media message originated by or received by the user via a computer networked social media service;

determining, by executing an instruction with the processor, first demographics associated with the first asset from a computer database based on a first asset identifier included in the computer database, the first asset identifier associated with the first asset, the first demographics including a first set of demographic categories, each demographic category including a likelihood of representing the first demographics;

in response to determining the first demographics associated with the first asset, associating, by executing an instruction with the processor, the first demographics with the user;

identifying, by executing an instruction with the processor, a second social media message of the asset-regarding social media messages mentioning a second asset, the second social media message originated by or received by the user via the computer networked social media service;

determining, by executing an instruction with the processor, second demographics associated with the second asset from the computer database based on a second asset identifier included in the computer database, the second asset identifier associated with the second asset, the second demographics including a second set of demographic categories, each demographic category including a likelihood of representing the second demographics;

tagging, by executing an instruction with the processor, a first user identifier with the first demographics and tagging a second user identifier with the second demographics;

based on the tagging, generating, by executing an instruction with the processor, a composite demographic profile for the user based on analyzing the likelihoods for each demographic category from the first demographics and the second demographics; and

storing, by executing an instruction with the processor, the composite demographic profile in a memory of a profile database to reduce storage requirements.

2. The method as defined in claim 1, wherein the identifying of the first social media message includes:

identifying a reference to the first asset in text of a social media message; and

determining whether a characteristic of the social media message satisfies a rule associated with the first asset.

3. The method as defined in claim 2, wherein the first asset is a media event and the rule associated with the first asset specifies a relationship to a broadcast time of the asset.

4. The method as defined in claim 2, wherein the determining of whether the characteristic of the social media message satisfies the rule associated with the first asset further includes determining whether a timestamp associated with the social media message falls within a time period defined by the rule.

5. The method as defined in claim 2, wherein the first asset is a product and the rule associated with the first asset identifies a vendor that supplies the first asset.

6. The method as defined in claim 5, wherein the determining of whether the characteristic of the social media message satisfies the rule associated with the first asset further includes determining whether the text of the social media message mentions the vendor.

7. The method as defined in claim 6, wherein the first demographics are associated with a group of people who are purchasers of the first asset.

8. The method as defined in claim 1, wherein the first demographics are associated with a group of people who are an audience for the first asset.

9. The method as defined in claim 1, wherein the first social media message and the second social media message are associated with a first user identifier, the method further including:

identifying a third social media message mentioning a third asset, the third social media message associated with a second user identifier associated with the user, the second user identifier different than the first user identifier; and

the estimating of the demographic profile for the user includes combining demographics associated with two or more different groups with which the first user identifier and the second user identifier are associated.

10. The method as defined in claim 1, wherein the first demographics are associated with a first group of people who are an audience for the first asset and the second demographics are associated with a second group of people who are purchasers of the second asset.

11. The method as defined in claim 1, wherein the estimating of the demographic profile for the user further includes:

analyzing the first demographics and the second demographics;

determining a demographic category with a highest likelihood based on results of the analyzing; and

associating the demographic category with the highest likelihood with the user.

12. The method as defined in claim 11, wherein the analyzing of the first demographics and the second demographics includes performing Bayesian analysis on the first demographics and the second demographics.

13. The method as defined in claim 11, wherein the analyzing of the first demographics and the second demographics includes performing principal component analysis on the first demographics and the second demographics.

14. A system comprising:

a tagged identifier logger to;

retrieve a plurality of electronically transmitted social media messages from a message hosting server based on media referenced in the plurality of electronically transmitted social media messages to reduce a set of social media messages for subsequent processing;

identify asset-regarding social media messages of the plurality of electronically transmitted social media messages based on references to assets in respective ones of the social media messages;

identify a first social media message of the asset-regarding social media messages mentioning a first asset, the first social media message originated by or received by a user via a computer networked social media service;

determine first demographics associated with the first asset from a computer database based on a first asset identifier included in the computer database, the first asset identifier associated with the first asset, the first demographics including a first set of demographic categories, each demographic category including a likelihood of representing the first demographics;

in response to determining the first demographics associated with the first asset, associate the first demographics with the user;

identify a second social media message of the asset-regarding social media messages mentioning a second asset, the second social media message originated by or received by the user via the computer networked social media service;

determine second demographics associated with the second asset from the computer database based on a second asset identifier included in the computer database, the second asset identifier associated with the second asset, the second demographics including a second set of demographic categories, each demographic category including a likelihood of representing the second demographics;

tag a first user identifier with the first demographics and tag a second user identifier with the second demographics; and

a profile generator to, based on the tagging, generate a composite demographic profile for the user based on analyzing the likelihoods for each demographic category from the first demographics and the second demographics and store the composite demographic profile in a memory of a profile database to reduce storage requirements,

at least one of the tagged identifier logger or the profile generator implemented by a logic circuit.

15. The system as defined in claim 14, further including:

a message analyzer to determine whether the first social media message mentions the first asset; and

a rules checker to determine whether a characteristic of the first social media message satisfies a rule associated with the first asset.

16. The system as defined in claim 15, wherein the first asset is a media event and the rule associated with the first asset specifies a relationship to a broadcast time of the first asset.

17. The system as defined in claim 16, wherein the rules checker is to determine whether the characteristic of the first social media message satisfies the rule associated with the first asset by determining whether a timestamp associated with the first social media message falls within a threshold time of the broadcast time of the first asset.

18. The system as defined in claim 16, wherein the first demographics are associated with a group of people who are an audience for the first asset.

19. The system as defined in claim 15, wherein the first asset is a product and the rule associated with the first asset specifies a vendor that supplies the first asset.

20. The system as defined in claim 19, wherein the rules checker is to determine whether the characteristic of the first social media message satisfies the rule associated with the first asset by determining whether the first social media message includes a reference to the vendor.

21. The system as defined in claim 19, wherein the first demographics are associated with a group of people who are purchasers of the first asset.

22. The system as defined in claim 14, wherein the first demographics are associated with a first group of people who are an audience for the first asset, and the second demographics are associated with a second group of people who are purchasers of the second asset.

23. The system as defined in claim 14, further including:

an analyzer to analyze the first demographics and the second demographics; and

an estimator to determine a demographic category with a highest likelihood based on an output of the analyzer, the estimator to associate the demographic category with the highest likelihood with the user.

24. The system as defined in claim 23, wherein the analyzer applies Bayesian analysis to the first demographics and the second demographics.

25. The system as defined in claim 23, wherein the analyzer applies principal component analysis to the first demographics and the second demographics.

26. A tangible computer readable storage medium comprising instructions that, when executed, cause a processor to at least:

retrieve a plurality of electronically transmitted social media messages from a message hosting server based on media referenced in the plurality of electronically transmitted social media messages to reduce a set of social media messages for subsequent processing;

identify asset-regarding social media messages of the plurality of electronically transmitted social media messages based on references to assets in respective ones of the social media messages

identify a first social media message of the asset-regarding social media messages mentioning a first asset, the first social media message originated by or received by a user via a computer networked social media service;

determine first demographics associated with the first asset from a computer database based on a first asset identifier included in the computer database, the first asset identifier associated with the first asset, the first demographics including a first set of demographic categories, each demographic category including a likelihood of representing the first demographics;

in response to determining the first demographics associated with the first asset, associate the first demographics with the user;

identify a second social media message of the asset-regarding social media messages mentioning a second asset, the second social media message originated by or received by the user via the computer networked social media service;

determine second demographics associated with the second asset from the computer database based on a second asset identifier included in the computer database, the second asset identifier associated with the second asset, the second demographics including a second set of demographic categories, each demographic category including a likelihood of representing the second demographics;

tag a first user identifier with the first demographics and tagging a second user identifier with the second demographics;

based on the tagging, generate a composite demographic profile for the user based on analyzing the likelihoods for each demographic category from the first demographics and the second demographics; and

store the composite demographic profile in a memory of a profile database to reduce storage requirements.

27. The tangible computer readable storage medium as defined in claim 26, wherein the instructions cause the processor to identify the first social media message by:

identifying a reference to the first asset in text of a social media message; and

determining whether a characteristic of the social media message satisfies a rule associated with the first asset.

28. The tangible computer readable storage medium as defined in claim 27, wherein the first asset is a media event and the rule specifies a timeframe after a broadcast time of the first asset.

29. The tangible computer readable storage medium as defined in claim 28, wherein the instructions cause the processor to determine whether the characteristic of the first social media message satisfies the rule by determining whether a timestamp associated with the first social media message occurs within the timeframe.

30. The tangible computer readable storage medium as defined in claim 27, wherein the first asset is a product and wherein the rule specifies a vendor that supplies the first asset.

31. The tangible computer readable storage medium as defined in claim 30, wherein the instructions further cause the processor to determine whether the text of the first social media message mentions the vendor.

32. The tangible computer readable storage medium as defined in claim 26, wherein the instructions further cause the processor to:

analyze the first demographics and the second demographics;

determine a demographic category with a highest likelihood based on results of the analysis; and

associate the demographic category with the highest likelihood with the user.

33. The tangible computer readable storage medium as defined in claim 32, wherein the instructions cause the processor to analyze the first demographics and the second demographics by performing Bayesian analysis on the first demographics and the second demographics.

34. The tangible computer readable storage medium as defined in claim 32, wherein the instructions cause the processor to analyze the first demographics and the second demographics by performing principal component analysis on the first demographics and the second demographics.