Message selection and presentation in a real-time messaging platform

• US 10,263,924 B1
• Filed: 10/30/2013
• Issued: 04/16/2019
• Est. Priority Date: 10/30/2013
• Status: Active Grant

• Current Assignee: Twitter Inc (X Holdings Corp.)
• Original Assignee: Twitter Inc (X Holdings Corp.)
• Inventors: Davidson, Michael, Brewer, Joshua, DeSandro, David, Nolan, Erin, Brown, Geoffrey
• Primary Examiner(s): Chou, Alan S
• Application Number:
  US14/067,061
• Time in Patent Office: 1,994 Days
• Field of Search: 709206
• CPC Class Codes: H04L 51/046 Interoperability with other network applications or services

Abstract

A system and message for message selection and presentation, including: receiving a request for content, wherein the request is associated with a context account of a messaging platform; identifying a set of unviewed messages associated with the context account; identifying grouping criteria for grouping content based on messages; applying, by a computer processor, the grouping criteria to the unviewed messages to generate a content group, wherein the content group is associated with a subset of the unviewed messages; and providing at least a portion of the content group in response to the request.

Claims

1. A method for message selection, comprising:

receiving a request for content, wherein the request is associated with a context account of a messaging platform;

identifying a set of unviewed messages in a message stream of the context account, wherein the set of unviewed messages were included in the message stream subsequent to the most recent viewing of the message stream by the context account, and wherein the unviewed messages are associated with accounts of the messaging platform having a predefined graph relationship with the context account;

identifying a plurality of distinct grouping criteria for grouping messages, wherein the plurality of grouping criteria include a popularity of a given message in the messaging platform and a relationship between the context account and an authoring account of the given message;

applying, by a computer processor, the plurality of grouping criteria to the unviewed messages to generate a plurality of content groups, each content group providing a grouping of unviewed messages of the messaging platform, wherein;

each content group is associated with a subset of the unviewed messages identified according to the particular grouping criteria for that content group, and

applying the grouping criteria for each generated content group comprises;

ranking the set of unviewed messages of the content group according to a plurality of message criteria, wherein each message criterion is assigned a corresponding weight to provide an overall ranking value to each unviewed message, and

selecting, based on ranking the set of unviewed messages, the subset of the unviewed messages for inclusion in the content group, wherein selecting comprises excluding at least one message of the set of unviewed messages from inclusion in the content group based on ranking the set of unviewed messages; and

providing, in response to the request, one or more of the plurality of content groups and in response to a user selection of a particular content group, providing at least a portion of the content group for display by a client device associated with the context account.

2. The method of claim 1, further comprising:

identifying favorite accounts of the context account;

selecting messages from the set of unviewed messages that are authored by the favorite accounts to form a group of messages;

ranking the group of messages based on a message criterion; and

selecting, based on ranking the group of messages, the subset of the unviewed messages from the group of messages for inclusion in the content group, wherein the subset of the unviewed messages is displayed by the client device.

3. The method of claim 2, wherein the favorite accounts are identified based on a favorite accounts criterion, and wherein the favorite accounts criterion is based on at least one selected from a group consisting of an indication of favorite accounts by the context account, a predefined graph relationship between the favorite accounts and the context account, and a frequency and recency of engagement between the favorite accounts and the context account.

4. The method of claim 2, wherein the message criterion is based on at least one selected from a group consisting of a recency and frequency of engagement with a message in the group of messages by a favorite account and geographical locations of a broadcasting user account of a message in the group of messages and the context account.

5. The method of claim 1, further comprising:

matching the set of unviewed messages with a set of topics to form a set of topic groups;

ranking the set of topic groups based on a topics criterion; and

selecting, based on ranking the set of topic groups, a subset of the topic groups for inclusion in the content group, wherein the subset of the topic groups are displayed with their matched messages by the client device.

6. The method of claim 5, further comprising dynamically adding and removing topics from the set of topics based at least on a recency and a frequency of keywords and phrases among messages broadcasted by the messaging platform.

7. The method of claim 5, wherein ranking the set of topic groups further comprises determining an average ranking score for each topic of the set of topic groups based on ranking scores for individual messages of the topic.

8. The method of claim 5, wherein selecting the subset of the topic groups comprises:

applying a weighted interest score for each topic of the set of topics, wherein the weighted interest score is an estimate of user interest in the topic.

9. The method of claim 5, wherein the topics criterion is based on at least one selected from a group consisting of a recency of viewing content associated with a topic, a frequency of viewing content associated with a topic, engagement with content associated with a topic, and a geographical proximity between the context account and the authoring account of content associated with a topic.

10. The method of claim 1, further comprising:

identifying a plurality of stories referenced by the set of unviewed messages;

ranking the plurality of stories based on a stories criterion; and

selecting the subset of the unviewed messages corresponding to a highest ranking story of the plurality of stories, wherein at least one of the subset of the unviewed messages is displayed by the client device concurrently with an indication of the highest ranking story.

11. The method of claim 10, further comprising:

selecting a description for the highest ranking story, wherein the indication of the highest ranking story comprises the description.

12. The method of claim 10, wherein the stories criterion is based on at least one selected from a group consisting of engagement with references to the plurality of stories by user accounts within a connection graph of the context account, and a popularity score of an authoring account of at least one unviewed message associated with at least one of the plurality of stories.

13. The method of claim 10, wherein the stories criterion is based on a count of references associated with each story.

14. The method of claim 10, wherein the reference is a uniform resource locator (URL) and the story is a web page referenced by the URL.

15. A system for message selection, comprising:

a computer processor;

a content-analysis module executing on the computer processor and configured to;

receive a request for content, wherein the request is associated with a context account;

identify a set of unviewed messages in a message stream of the context account, wherein the set of unviewed messages were included in the message stream subsequent to the most recent viewing of the message stream by the context account, and wherein the unviewed messages are associated with accounts of the messaging platform having a predefined graph relationship with the context account;

identify a plurality of distinct grouping criteria for grouping messages, wherein the plurality of grouping criteria include a popularity of a given message in the messaging platform and a relationship between the context account and an authoring account of the given message;

apply the plurality of grouping criteria to the unviewed messages to generate a plurality of content groups, each content group providing a grouping of unviewed messages of the messaging platform, wherein;

each content group is associated with a subset of the unviewed messages identified according to the particular grouping criteria for that content group, and

applying the grouping criteria for each generated content group comprises;

ranking the set of unviewed messages of the content group according to a plurality of message criteria, wherein each message criterion is assigned a corresponding weight to provide an overall ranking value to each unviewed message, and

selecting, based on ranking the set of unviewed messages, the subset of the unviewed messages for inclusion in the content group, wherein selecting comprises excluding at least one message of the set of unviewed messages from inclusion in the content group based on ranking the set of unviewed messages; and

provide, in response to the request, one or more of the plurality of content groups and in response to a user selection of a particular content group, providing at least a portion of the content group for display by a client device associated with the context account.

16. The system of claim 15, wherein the content-analysis module is further configured to:

identify favorite accounts of the context account;

select messages from the set of unviewed messages that are authored by the favorite accounts to form a group of messages;

rank the group of messages based on a message criterion; and

select, based on ranking the group of messages, the subset of the unviewed messages from the group of messages for inclusion in the content group, wherein the subset of the unviewed messages is displayed by the client device.

17. The system of claim 16, wherein the favorite accounts are identified based on a favorite accounts criterion, and wherein the favorite accounts criterion is based on at least one selected from a group consisting of an indication of favorite accounts by the context account, a predefined graph relationship between the favorite accounts and the context account, and a frequency and recency of engagement between the favorite accounts and the context account.

18. The system of claim 16, wherein the message criterion is based on at least one selected from a group consisting of a recency and frequency of engagement with a message in the group of messages by a favorite account and geographical locations of a broadcasting user account of a message in the group of messages and the context account.

19. The system of claim 15, wherein the content-analysis module is further configured to:

match the set of unviewed messages with a set of topics to form a set of topic groups;

rank the set of topic groups based on a topics criterion; and

select, based on ranking the set of topic groups, a subset of the topic groups for inclusion in the content group, wherein the subset of the topic groups are displayed with their matched messages by the client device.

20. The system of claim 19, wherein the content-analysis module is further configured to dynamically add and removing topics from the set of topics based at least on a recency and a frequency of keywords and phrases among messages broadcasted by the messaging platform.

21. The system of claim 19, wherein ranking the set of topic groups further comprises determining an average ranking score for each topic of the set of topic groups based on ranking scores for individual messages of the topic.

22. The system of claim 19, wherein selecting the subset of the topic groups comprises:

applying a weighted interest score for each topic of the set of topics, wherein the weighted interest score is an estimate of user interest in the topic.

23. The system of claim 19, wherein the topics criterion is based on at least one selected from a group consisting of a recency of viewing content associated with a topic, a frequency of viewing content associated with a topic, engagement with content associated with a topic, and a geographical proximity between the context account and the authoring account of content associated with a topic.

24. The system of claim 15, wherein the content-analysis module is further configured to:

identify a plurality of stories referenced by the set of unviewed messages;

rank the plurality of stories based on a stories criterion; and

select the subset of the unviewed messages corresponding to a highest ranking story of the plurality of stories, wherein at least one of the subset of the unviewed messages is displayed by the client device concurrently with an indication of the highest ranking story.

25. The system of claim 24, wherein the content-analysis module is further configured to:

select a description for the highest ranking story, wherein the indication of the highest ranking story comprises the description.

26. The system of claim 24, wherein the stories criterion is based on at least one selected from a group consisting of engagement with references to the plurality of stories by user accounts within a connection graph of the context account, and a popularity score of an authoring account of at least one unviewed message associated with at least one of the plurality of stories.

27. The system of claim 24, wherein the stories criterion is based on a count of references associated with each story.

28. The system of claim 24, wherein the reference is a uniform resource locator (URL) and the story is a web page referenced by the URL.

29. A computer program product, encoded on one or more non-transitory computer-readable media, comprising software that when executed by one or more computers cause the one or more computers to perform operations comprising:

receiving a request for content, wherein the request is associated with a context account of a messaging platform;

identifying a set of unviewed messages in a message stream of the context account, wherein the set of unviewed messages were included in the message stream subsequent to the most recent viewing of the message stream by the context account, and wherein the unviewed messages are associated with accounts of the messaging platform having a predefined graph relationship with the context account;

identifying a plurality of distinct grouping criteria for grouping messages, wherein the plurality of grouping criteria include a popularity of a given message in the messaging platform and a relationship between the context account and an authoring account of the given message;

applying, by a computer processor, the plurality of grouping criteria to the unviewed messages to generate a plurality of content groups, each content group providing a grouping of unviewed messages of the messaging platform, wherein;

each content group is associated with a subset of the unviewed messages identified according to the particular grouping criteria for that content group, and

applying the grouping criteria for each generated content group comprises;

ranking the set of unviewed messages of the content group according to a plurality of message criteria, wherein each message criterion is assigned a corresponding weight to provide an overall ranking value to each unviewed message, and

selecting, based on ranking the set of unviewed messages, the subset of the unviewed messages for inclusion in the content group, wherein selecting comprises excluding at least one message of the set of unviewed messages from inclusion in the content group based on ranking the set of unviewed messages; and

providing, in response to the request, one or more of the plurality of content groups and in response to a user selection of a particular content group, providing at least a portion of the content group for display by a client device associated with the context account.

30. The computer program product of claim 29, further comprising software that when executed cause the one or more computers to perform operations comprising:

identifying favorite accounts of the context account;

selecting messages from the set of unviewed messages that are authored by the favorite accounts to form a group of messages;

ranking the group of messages based on a message criterion; and

selecting, based on ranking the group of messages, the subset of the unviewed messages from the group of messages for inclusion in the content group, wherein the subset of the unviewed messages is displayed by the client device.

31. The computer program product of claim 30, wherein the favorite accounts are identified based on a favorite accounts criterion, and wherein the favorite accounts criterion is based on at least one selected from a group consisting of an indication of favorite accounts by the context account, a predefined graph relationship between the favorite accounts and the context account, and a frequency and recency of engagement between the favorite accounts and the context account.

32. The computer program product of claim 30, wherein the message criterion is based on at least one selected from a group consisting of a recency and frequency of engagement with a message in the group of messages by a favorite account and geographical locations of a broadcasting user account of a message in the group of messages and the context account.

33. The computer program product of claim 29, further comprising software that when executed cause the one or more computers to perform operations comprising:

matching the set of unviewed messages with a set of topics to form a set of topic groups;

ranking the set of topic groups based on a topics criterion; and

selecting, based on ranking the set of topic groups, a subset of the topic groups for inclusion in the content group, wherein the subset of the topic groups are displayed with their matched messages by the client device.

34. The computer program product of claim 33, further comprising software that when executed cause the one or more computers to perform operations comprising dynamically adding and removing topics from the set of topics based at least on a recency and a frequency of keywords and phrases among messages broadcasted by the messaging platform.

35. The computer program product of claim 33, wherein ranking the set of topic groups further comprises determining an average ranking score for each topic of the set of topic groups based on ranking scores for individual messages of the topic.

36. The computer program product of claim 33, wherein selecting the subset of the topic groups comprises:

applying a weighted interest score for each topic of the set of topics, wherein the weighted interest score is an estimate of user interest in the topic.

37. The computer program product of claim 33, wherein the topics criterion is based on at least one selected from a group consisting of a recency of viewing content associated with a topic, a frequency of viewing content associated with a topic, engagement with content associated with a topic, and a geographical proximity between the context account and the authoring account of content associated with a topic.

38. The computer program product of claim 29, further comprising software that when executed cause the one or more computers to perform operations comprising:

identifying a plurality of stories referenced by the set of unviewed messages;

ranking the plurality of stories based on a stories criterion; and

selecting the subset of the unviewed messages corresponding to a highest ranking story of the plurality of stories, wherein at least one of the subset of the unviewed messages is displayed by the client device concurrently with an indication of the highest ranking story.

39. The computer program product of claim 38, further comprising software that when executed cause the one or more computers to perform operations comprising:

selecting a description for the highest ranking story, wherein the indication of the highest ranking story comprises the description.

40. The computer program product of claim 38, wherein the stories criterion is based on at least one selected from a group consisting of engagement with references to the plurality of stories by user accounts within a connection graph of the context account, and a popularity score of an authoring account of at least one unviewed message associated with at least one of the plurality of stories.

41. The computer program product of claim 38, wherein the stories criterion is based on a count of references associated with each story.

42. The computer program product of claim 38, wherein the reference is a uniform resource locator (URL) and the story is a web page referenced by the URL.