Confusion reduction in an online social network

• US 10,397,168 B2
• Filed: 08/31/2016
• Issued: 08/27/2019
• Est. Priority Date: 08/31/2016
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bastide, Paul R., Broomhall, Matthew E., Loredo, Robert E.
• Primary Examiner(s): Adams, Charles D
• Application Number:
  US15/252,827
• Publication Number:
  US 20180060416A1
• Time in Patent Office: 1,091 Days
• Field of Search: None
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

Confusion reduction in an online social network. A method identifies elements in social media message content, determines, based on the identified elements, a plurality of candidate audiences to which the social media message content is potentially targeted, indicates, to a user, the plurality of candidate audiences and suggested additional elements to apply to the social media message content to tailor the social media message content to a target audience of the plurality of candidate audiences, and modifies the social media message content with one or more additional elements of the suggested additional elements, the modifying adding the one or more additional elements to the social media message content and targeting the social media message content to the target audience.

Claims

1. A computer-implemented method comprising:

identifying elements in social media message content, the social media message content comprising a posted message posted to a social media platform;

determining whether the social media message content is indefinite as to an audience being targeted, the determining whether the social media message content is indefinite as to an audience being targeted comprising predicting a likelihood of confusion based on the social media message content, wherein the prediction of the likelihood of confusion is based on an age of the posted message, in which the older the post, the higher the predicted likelihood of confusion;

determining, based on the identified elements, a plurality of different candidate audiences to which the social media message content is potentially targeted, each candidate audience of the plurality of difference candidate audiences ascertained based on a respective corresponding contextual understanding, of a plurality of different contextual understandings, given to the social media message content, wherein the determining the plurality of different candidate audiences comprises;

building a respective dictionary for each user of a plurality of users of a social media platform in which the social media message content is composed, wherein a dictionary for a given user of the plurality of users comprises elements include in prior-composed social media messages composed by the given user;

ascertaining a frequency of the elements included in prior-composed social media messages composed by each user;

building a clustered representation of the social media platform using k-means against the frequency of the elements;

querying a message space for social media messages based on the social media message content; and

identifying dense k-clusters based on the social media message content, the dense k-clusters corresponding to the plurality of different candidate audiences;

indicating to a user the plurality of candidate audiences and, for each candidate audience of the plurality of different candidate audiences, a suggested one or more additional elements to apply to the social media message content to provide additional context for the social media message content and thereby tailor the social media message content to an audience of the plurality of different candidate audiences and corresponding contextual understanding; and

modifying the social media message content with the one or more additional elements for a target audience of the plurality of different candidate audiences, the modifying adding the one or more additional elements to the social media message content and targeting the social media message content to the target audience.

2. The method of claim 1, further comprising:

identifying a communications channel identified in the social media message content; and

obtaining at least some of the suggested additional elements from other messages identifying the communications channel.

3. The method of claim 1, further comprising:

presenting an interface through which the user selects the one or more additional elements with which the social media message content is modified, the interface presenting selectable indications of the plurality of different candidate audiences, and, for each audience of the plurality of different candidate audiences, the respective one or more additional elements to augment the social media message content to target the social media message content;

receiving via the interface user selections of the one or more additional elements with which to modify the social media message content; and

performing the modifying based on receiving the user selections, wherein the social media message content is augmented with the user selections of the one or more additional elements.

4. The method of claim 1, wherein the identified elements comprise natural language elements and one or more metadata tags of the social media message content.

5. The method of claim 1, wherein at least one of the respective one or more additional elements for a candidate audience comprises a metadata tag.

6. The method of claim 1, wherein the social media message content is selected from the group consisting of:

a social media post, a comment to a social media post, a message between users in a social media platform, and a forum post.

7. The method of claim 1, wherein the social media message content comprises a message not yet posted to a social media platform for which the social media message content is composed, and wherein the method performs the modifying the social media message content prior to posting the social media message content as the message to the social media platform.

8. A computer program product comprising:

a computer readable storage medium readable by a processing circuit and storing instructions for execution by the processing circuit for performing a method comprising;

identifying elements in social media message content, the social media message content comprising a posted message posted to a social media platform;

determining whether the social media message content is indefinite as to an audience being targeted, the determining whether the social media message content is indefinite as to an audience being targeted comprising predicting a likelihood of confusion based on the social media message content, wherein the prediction of the likelihood of confusion is based on an age of the posted message, in which the older post, the higher the predicted likelihood of confusion;

determining, based on the identified elements, a plurality of different candidate audiences to which the social media message content is potentially targeted, each candidate audience of the plurality of difference candidate audiences ascertained based on a respective corresponding contextual understanding, of a plurality of different contextual understandings, given to the social media message content, wherein the determining the plurality of different candidate audiences comprises;

building a respective dictionary for each user of a plurality of users of a social media platform in which the social media message content is composed, wherein a dictionary for a given user of the plurality of users comprises elements included in prior-composed social media messages composed by the given user;

ascertaining a frequency of the elements included in prior-composed social media message composed by each user;

building a clustered representation of the social media platform using k-means against the frequency of the elements;

querying a message space for social media messages based on the social media message content; and

identifying dense k-clusters based on the social media message content, the dense k-clusters corresponding to the plurality of different candidate audiences;

indicating to a user the plurality of candidate audiences and, for each candidate audience of the plurality of different candidate audiences, a suggested one or more additional elements to apply to the social media message content to provide additional context for the social media message content and thereby tailor the social media message content to an audience of the plurality of different candidate audiences and corresponding contextual understanding; and

modifying the social media message content with the one or more additional elements for a target audience of the plurality of different candidate audiences, the modifying adding the one or more additional elements to the social media message content and targeting the social media message content to the target audience.

9. The computer program product of claim 8, wherein the method further comprises:

identifying a communications channel identified in the social media message content; and

obtaining at least some of the suggested additional elements from other messages identifying the communications channel.

10. The computer program product of claim 8, wherein the social media message content comprises a message not yet posted to a social media platform for which the social media message content is composed, and wherein the method performs the modifying the social media message content prior to posting the social media message content as the message to the social media platform.

11. A computer system comprising:

a memory; and

a processor in communication with the memory, wherein the computer system is configured to perform a method, the method comprising;

identifying elements in social media message content, the social media message content comprising a posted message posted to a social media platform;

determining whether the social message content is indefinite as to an audience being targeted, the determining whether the social media message content is indefinite as to an audience being targeted comprising predicting a likelihood of confusion based on the social media message content, wherein the prediction of the likelihood of confusion is based on an age of the posted message, in which the older post, the higher the predicted likelihood of confusion;

determining, based on the identified elements, a plurality of different candidate audiences to which the social media message content is potentially targeted, each candidate audience of the plurality of difference candidate audiences ascertained based on a respective corresponding contextual understanding, of a plurality of different contextual understandings, given to the social media message content, wherein the determining the plurality of different candidate audiences comprises;

building a respective dictionary for each user of a plurality of users of a social media platform in which the social media message content is composed, wherein a dictionary for a given user of the plurality of users comprises elements included in prior-composed social media messages composed by the given user;

ascertaining a frequency of the elements included in prior-composed social media messages composed by each user;

building a clustered representation of the social media platform using k-means against the frequency of the elements;

querying a message space for social media messages based on the social media message content; and

identifying dense k-clusters based on the social media message content, the dense k-clusters corresponding to the plurality of different candidate audiences;

indicating to a user the plurality of candidate audiences and, for each candidate audience of the plurality of different candidate audiences, a suggested one or more additional elements to apply to the social media message content to provide additional context for the social media message content and thereby tailor the social media message content to an audience of the plurality of different candidate audiences and corresponding contextual understanding; and

modifying the social media message content with the one or more additional elements for a target audience of the plurality of different candidate audiences, the modifying adding the one or more additional elements to the social media message content and targeting the social media message content to the target audience.

12. The computer system of claim 11, wherein the method further comprises:

identifying a communications channel identified in the social media message content; and

obtaining at least some of the suggested additional elements from other messages identifying the communications channel.

13. The computer system of claim 11, wherein the social media message content comprises a message not yet posted to a social media platform for which the social media message content is composed, and wherein the method performs the modifying the social media message content prior to posting the social media message content as the message to the social media platform.