User profile stitching

• US 9,710,555 B2
• Filed: 02/25/2011
• Issued: 07/18/2017
• Est. Priority Date: 05/28/2010
• Status: Active Grant

• Current Assignee: Adobe Inc.
• Original Assignee: Adobe Systems Incorporated (Adobe Inc.)
• Inventors: Smith, Kevin G., George, William Brandon
• Primary Examiner(s): Waldron, Scott A
• Assistant Examiner(s): Ohba, Mellissa M
• Application Number:
  US13/035,286
• Publication Number:
  US 20150293997A1
• Time in Patent Office: 2,335 Days
• CPC Class Codes: G06F 16/95 Retrieval from the web
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0269 based on user profile or attribute
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

Methods and techniques may be provided for matching user profiles on distinct social media platforms. A first profile is retrieved from a first social media platform. The first profile is identified by one or more search criteria. Responsive to identifying an instance of the one or more search criteria on a second social media platform, a first probability is determined. A second profile is identified by the one or more search criteria on the second social media platform. The first probability predicts whether the second profile is associated with an entity associated with the first profile. The determining the first probability further comprises comparing an attribute of the first profile to an attribute of the second profile. Responsive to determining that the first probability exceeds a threshold, a match indicating that the second profile is associated with an entity associated with the first profile is recorded.

Claims

1. A method for determining that multiple social media profiles correspond to a common entity, the method comprising:

identifying, by a social media analysis tool executed by a processor, a first profile for a first social media platform, a second profile for a second social media platform, and a third profile for a third social media platform, wherein the first and the third profile have a first attribute, a second attribute, and a third attribute, wherein the second profile has the first attribute and lacks the second and third attributes, wherein the second attribute describes an entity name and the third attribute describes a geographical location;

determining, by the social media analysis tool based on the first profile and the second profile having different values for the first attribute, that a probability of both the first profile and the second profile corresponding to the common entity is less than a threshold;

determining, by the social media analysis tool, a uniqueness score for a combination of values of the second attribute and the third attribute in the first profile, wherein the uniqueness score indicates a likelihood of the second and third attributes having the combination of values, wherein determining the uniqueness score comprises determining a uniqueness of a name identified by the entity name from the second attribute being associated with the geographical location from the third attribute;

selecting, by the social media analysis tool based on the determined uniqueness score, the second and third attributes as a basis of comparison between the first profile and the third profile;

matching, by the social media analysis tool, the first profile to the third profile based on comparing the values in the first profile for the selected second and third attributes with values in the third profile for the selected second and third attributes;

matching, by the social media analysis tool, the second profile and the third profile based at least in part on the second and third profiles having corresponding values for the first attribute; and

updating, by the social media analysis tool, the probability of both the first profile and the second profile corresponding to the common entity so that the probability exceeds the threshold, wherein the probability is updated based on matching the first profile to the third profile using the second and third attributes and matching the second profile to the third profile using the first attribute.

2. The method of claim 1, further comprising determining that the uniqueness score exceeds a threshold uniqueness score, wherein selecting the second and third attributes as the basis of comparison between the first profile and the third profile is performed based at least in part on the uniqueness score exceeding the threshold uniqueness score.

3. The method of claim 1, wherein matching the first profile to the third profile comprises comparing a web page attribute for the first profile with a web page attribute for the third profile.

4. The method of claim 3, wherein each of the web page attribute for the first profile and the web page attribute for the third profile describes, for each profile of the first profile and the third profile, at least one of:

a respective web page linked by the profile and external to each of the first social media platform and the third social media platform; and

a respective web page linked by a social media thread modified by or accessed by the profile and external to each of the first social media platform and the third social media platform.

5. The method of claim 3, wherein each of the web page attribute for the first profile and the web page attribute for the third profile describes a respective word frequency distribution for a respective web page accessible via the profile that is external to each of the first social media platform and the third social media platform.

6. The method of claim 1, wherein matching the first profile to the third profile further comprises comparing a social media content attribute for the first profile with a social media content attribute for the third profile.

7. The method of claim 6, wherein each of the social media content attribute for the first profile and the social media content attribute for the third profile describes, for each profile of the first profile and the third profile, at least one of:

a respective social media post authored by the profile and accessible via the first social media platform or the second social media platform;

a respective social media post rated by the profile and accessible via the first social media platform or the second social media platform; and

a respective social media post flagged by the profile and accessible via the first social media platform or the second social media platform.

8. The method of claim 6, wherein each of the social media content attribute for the first profile and the social media content attribute for the third profile describes, for each profile of the first profile and the third profile, a respective word frequency distribution for a respective social media post accessible via the first social media platform or the third social media platform.

9. The method of claim 6, wherein each of the social media content attribute for the first profile and the social media content attribute for the third profile describes, for each profile of the first profile and the third profile, a respective writing style for a respective social media post accessible via the first social media platform or the third social media platform.

10. The method of claim 1, further comprising:

prior to receiving the first profile;

receiving, from a client associated with the first profile, a search request for content from a content provider,

determining that the content provider lacks information regarding interactions with the client, and

responsive to determining that the content provider lacks the information regarding interactions with the client, requesting the first profile;

generating, subsequent to matching the first profile to the third profile and updating the probability, a recommendation for additional content based on determining that the first profile, the second profile, and third profile all correspond to the common entity; and

providing the recommendation to the client.

11. The method of claim 1, wherein the first attribute is a profile picture associated with a social media profile.

12. The method of claim 1, wherein the matching of the first profile to the third profile is used to tailor product recommendations to the user.

13. A non-transitory computer-readable medium embodying program code executable by a processor to provide a social media analysis tool for determining that multiple social media profiles correspond to a common entity, the program code comprising:

program code for identifying a first profile for a first social media platform, a second profile for a second social media platform, and a third profile for a third social media platform, wherein the first and the third profile have a first attribute, a second attribute, and a third attribute, wherein the second profile has the first attribute and lacks the second and third attributes, wherein the second attribute describes an entity name and the third attribute describes a geographical location;

program code for determining, based on the first profile and the second profile having different values for the first attribute, that a probability of both the first profile and the second profile corresponding to the common entity is less than a threshold;

program code for determining a uniqueness score for a combination of values of the second attribute and the third attribute in the first profile, wherein the uniqueness score indicates a likelihood of the second and third attributes having the combination of values, wherein determining the uniqueness score comprises determining a uniqueness of a name identified by the entity name from the second attribute being associated with the geographical location from the third attribute;

program code for selecting, based on the determined uniqueness score, the second and third attributes as a basis of comparison between the first profile and the third profile;

program code for matching the first profile to the third profile based on comparing the values in the first profile for the selected second and third attributes with values in the third profile for the selected second and third attributes;

program code for matching the second profile and the third profile based at least in part on the second and third profiles having corresponding values for the first attribute; and

program code for updating the probability of both the first profile and the second profile corresponding to the common entity so that the probability exceeds the threshold, wherein the probability is updated based on matching the first profile to the third profile using the second and third attributes and matching the second profile to the third profile using the first attribute.

14. The non-transitory computer-readable medium of claim 13, wherein the program code further comprises program code for determining that the uniqueness score exceeds a threshold uniqueness score, wherein selecting the second and third attributes as the basis of comparison between the first profile and the third profile is performed based at least in part on the uniqueness score exceeding the threshold uniqueness score.

15. The non-transitory computer-readable medium of claim 13, wherein the program code for matching the first profile to the third profile comprises program code for comparing a web page attribute for the first profile with a web page attribute for the third profile.

16. The non-transitory computer-readable medium of claim 15, wherein each of the web page attribute for the first profile and the web page attribute for the third profile describes, for each profile of the first profile and the third profile, at least one of:

a respective web page linked by the profile and external to each of the first social media platform and the third social media platform; and

a respective web page linked by a social media thread modified by or accessed by the profile and external to each of the first social media platform and the third social media platform.

17. The non-transitory computer-readable medium of claim 13, wherein the program code for wherein matching the first profile to the third profile comprises program code for comparing a social media content attribute for the first profile with a social media content attribute for the third profile,

wherein each of the social media content attribute for the first profile and the social media content attribute for the third profile describes, for each profile of the first profile and the third profile, at least one of:

a respective social media post authored by the profile and accessible via the first social media platform or the second social media platform;

a respective social media post rated by the profile and accessible via the first social media platform or the second social media platform; and

a respective social media post flagged by the profile and accessible via the first social media platform or the second social media platform.

18. A system comprising:

a processor, and

a non-transitory computer-readable medium communicatively coupled to the processor and embodying a social media analysis tool for determining that multiple social media profiles correspond to a common entity;

wherein the processor is configured to execute the social media analysis tool, wherein executing the social media analysis tool causes the processor to perform operations comprising;

identifying a first profile for a first social media platform, a second profile for a second social media platform, and a third profile for a third social media platform, wherein the first and the third profile have a first attribute, a second attribute, and a third attribute, wherein the second profile has the first attribute and lacks the second and third attributes, wherein the second attribute describes an entity name and the third attribute describes a geographical location,

determining, based on the first profile and the second profile having different values for the first attribute, that a probability of both the first profile and the second profile corresponding to the common entity is less than a threshold,

determining a uniqueness score for a combination of values of the second attribute and the third attribute in the first profile, wherein the uniqueness score indicates a likelihood of the second and third attributes having the combination of values, wherein determining the uniqueness score comprises determining a uniqueness of a name identified by the entity name from the second attribute being associated with the geographical location from the third attribute,

selecting, based on the determined uniqueness score, the second and third attributes as a basis of comparison between the first profile and the third profile,

matching the first profile to the third profile based on comparing the values in the first profile for the selected second and third attributes with values in the third profile for the selected second and third attributes,

matching the second profile and the third profile based at least in part on the second and third profiles having corresponding values for the first attribute, and

updating the probability of both the first profile and the second profile corresponding to the common entity so that the probability exceeds the threshold, wherein the probability is updated based on matching the first profile to the third profile using the second and third attributes and matching the second profile to the third profile using the first attribute.