Trends in a messaging platform

• US 10,601,749 B1
• Filed: 07/11/2014
• Issued: 03/24/2020
• Est. Priority Date: 07/11/2014
• Status: Active Grant

• Current Assignee: X Corp. (f/k/a Twitter, Inc.) (X Holdings Corp.)
• Original Assignee: X Corp. (f/k/a Twitter, Inc.) (X Holdings Corp.)
• Inventors: Snow, Rion Langley, Mishne, Gilad Avraham
• Primary Examiner(s): Bengzon, Greg C
• Application Number:
  US14/329,782
• Time in Patent Office: 2,083 Days
• CPC Class Codes: H04L 51/212 using filtering or selective blocking
• CPC Class Codes: H04L 51/214 using selective forwarding
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

A method of operating a messaging platform, including: obtaining, for a first profile of the messaging platform, a trending entity within a first topic of interest to the first profile; accessing a first plurality of messages classified as the first topic of interest; retrieving a subset of the first plurality of messages corresponding to the trending entity; and sending content associated with the subset for display to a user of the first profile.

Claims

1. A method comprising:

determining a plurality of topics of interest for a profile of a messaging platform,

each topic being associated with a respective intra-profile weight that is a measurement of a degree of interest of the topic for the profile;

computing, from entities referenced in messages authored by a plurality of users of the messaging platform, a respective plurality of trending entities for each topic of the plurality of topics, including;

computing, for each topic, a respective intra-topic weight for each entity belonging to the topic, the intra-topic weight for each entity belonging to a particular topic representing a significance of (i) a recent frequency for the entity with respect to the particular topic compared to (ii) a baseline frequency for the entity with respect to the particular topic, and

designating, for each topic, one or more entities having an intra-topic weight that satisfies a threshold as trending entities for the topic;

receiving, from a client device, a request for trending content to be provided to a user device associated with the profile of a messaging platform;

computing a profile-specific score for each trending entity of a plurality of trending entities belonging to the topics of interest determined for the profile, including multiplying a respective intra-topic weight for the trending entity by an intra-profile weight of a topic of the trending entity;

selecting, for each topic, a respective number of trending entities based on the respective profile-specific scores computed for the plurality of trending entities belonging to the topics of interest of the profile;

retrieving one or more messages associated with the selected trending entities; and

sending, to the client device, content associated with one or more of the retrieved messages for display to the user of the profile in a message stream, thereby reducing a likelihood of the user missing content that reflects trending entities in topics that are of interest to the user.

2. The method of claim 1, wherein retrieving the one or more messages associated with the selected trending entities comprises:

filtering messages associated with the selected trending entities according to a language preference of the profile to obtain a filtered set of messages,

wherein the one or more messages associated with the one or more selected trending entities are obtained from the filtered set of messages.

3. The method of claim 1, wherein retrieving the one or more messages associated with the selected trending entities comprises:

filtering messages associated with the selected trending entities according to a geographic region affiliated with the profile to obtain a filtered set of messages,

wherein the one or more messages associated with the one or more selected trending entities are obtained from the filtered set of messages.

4. The method of claim 1, wherein the content comprises a message in the one or more messages.

5. The method of claim 1, further comprising:

identifying, by parsing a message in the one or more messages, an image in the message and a source of the image;

comparing the source of the image against a plurality of whitelisted domains; and

adding, in response to determining that the source of the image is in one of the plurality of whitelisted domains, the image to the content for display to the user of the profile.

6. The method of claim 1, further comprising:

identifying, by parsing a message in the one or more messages, a link in the message to a news story and a source of the news story;

comparing the source of the news story against a plurality of whitelisted domains; and

adding, in response to determining that the source of the news story is in the plurality of whitelisted domains, the link to the content for display to the user of the profile.

7. The method of claim 1, wherein the recent frequency of a particular trending entity represents the occurrences of the particular trending entity in messages of the messaging platform over a specified time period.

8. The method of claim 1, wherein the profile is a first profile, and the intra-profile weight for a particular topic of interest is determined based on (i) a number of messages of the messaging platform that are associated with the particular topic and are issued or forwarded by the first profile, or (ii) a number of other profiles that are followed by the first profile and are known for issuing messages associated with the particular topic of interest.

9. The method of claim 1, wherein a particular trending entity is associated with a first topic and a second topic of the plurality of topics of interest, and wherein computing a profile-specific score for the particular trending entity includes calculating a combined weight for the particular trending entity by summing (i) a first profile-specific score of the particular trending entity for the first topic, and (ii) a second profile-specific score of the particular trending entity for the second topic.

10. The method of claim 1, further comprising:

for each topic, calculating a list size, the list size setting a maximum number (N) of trending entities to be selected for the topic, wherein the trending entities selected for the topic comprise N top trending entities associated with a top N intra-topic weights with respect to the topic.

11. The method of claim 10, wherein the list size is calculated by multiplying (i) an intra-profile weight of the topic with respect to the profile by (ii) a total number of trending entities to be selected for all topics of interest associated with the profile.

12. The method of claim 1, wherein a first entity belongs to a first topic and to a second topic different from the first topic, the first entity having a first intra-topic weight associated with the first topic, and a second intra-topic weight associated with the second topic.

13. The method of claim 1, wherein each of the selected trending entities is associated with a respective profile-specific score with respect to the profile, and each of the respective profile-specific scores of the selected entities is greater than a threshold value.

14. A non-transitory computer readable medium (CRM) storing instructions for operating a messaging platform, the instructions comprising:

determining a plurality of topics of interest for a profile of a messaging platform,

each topic being associated with a respective intra-profile weight that is a measurement of a degree of interest of the topic for the profile;

computing, from entities referenced in messages authored by a plurality of users of the messaging platform, a respective plurality of trending entities for each topic of the plurality of topics, including;

computing, for each topic, a respective intra-topic weight for each entity belonging to the topic, the intra-topic weight for each entity belonging to a particular topic representing a significance of (i) a recent frequency for the entity with respect to the particular topic, compared to (ii) a baseline frequency for the entity with respect to the particular topic, and

designating, for each topic, one or more entities having an intra-topic weight that satisfies a threshold as trending entities for the topic;

receiving, from a client device, a request for trending content to be provided to a user device associated with the profile of a messaging platform;

computing a profile-specific score for each trending entity of a plurality of trending entities belonging to the topics of interest determined for the profile, including multiplying a respective intra-topic weight for the trending entity by an intra-profile weight of a topic of the trending entity;

selecting, for each topic, a respective number of trending entities based on the respective profile-specific scores computed for the plurality of trending entities belonging to the topics of interest of the profile;

retrieving one or more messages associated with the selected trending entities; and

sending, to the client device, content associated with one or more of the retrieved messages for display to the user of the profile in a message stream, thereby reducing a likelihood of the user missing content that reflects trending entities in topics that are of interest to the user.

15. The non-transitory CRM of claim 14, wherein retrieving the one or more messages associated with the selected trending entities comprises:

filtering messages associated with the selected trending entities according to a language preference of the profile to obtain a filtered set of messages,

wherein the one or more messages associated with the one or more selected trending entities are obtained from the filtered set of messages.

16. The non-transitory CRM of claim 14, wherein retrieving the messages associated with the selected trending entities comprises:

filtering messages associated with the selected trending entities according to a geographic region affiliated with the profile to obtain a filtered set of messages,

wherein the one or more messages associated with the one or more selected trending entities are obtained from the filtered set of messages.

17. The non-transitory CRM of claim 14, wherein the content comprises a message in the one or more messages.

18. The non-transitory CRM of claim 14, the instructions further comprising:

identifying, by parsing a message in the one or more messages, an image in the message and a source of the image;

comparing the source of the image against a plurality of whitelisted domains; and

adding, in response to determining that the source of the image is in one of the plurality of whitelisted domains, the image to the content for display to the user of the profile.

19. The non-transitory CRM of claim 14, the instructions further comprising:

identifying, by parsing a message in the one or more messages, a link in the message to a news story and a source of the news story;

comparing the source of the news story against a plurality of whitelisted domains; and

adding, in response to determining that the source of the news story is in the plurality of whitelisted domains, the link to the content for display to the user of the profile.

20. The CRM of claim 14, wherein the profile is a first profile, and the intra-profile weight for a particular topic of interest is determined based on (i) a number of messages of the messaging platform that are associated with the particular topic and are issued or forwarded by the first profile, or (ii) a number of other profiles that are followed by the first profile and are known for issuing messages associated with the particular topic of interest.

21. The CRM of claim 14, wherein a particular trending entity is associated with a first topic and a second topic of the plurality of topics of interest, and wherein computing a profile-specific score for the particular trending entity includes calculating a combined weight for the particular trending entity by summing (i) a first profile-specific score of the particular trending entity for the first topic, and (ii) a second profile-specific score of the particular trending entity for the second topic.

22. The CRM of claim 14, wherein the instructions further comprise:

for each topic, calculating a list size, the list size setting a maximum number (N) of trending entities to be selected for the topic, wherein the trending entities selected for the topic comprise N top trending entities associated with a top N intra-topic weights with respect to the topic.

23. The CRM of claim 14, wherein a first entity belongs to a first topic and to a second topic different from the first topic, the first entity having a first intra-topic weight associated with the first topic, and a second intra-topic weight associated with the second topic.

24. A messaging platform system comprising:

one or more computers including a processor and a memory, the one or more computers configured to perform operations comprising;

determining a plurality of topics of interest for a profile of a messaging platform,

each topic being associated with a respective intra-profile weight that is a measurement of a degree of interest of the topic for the profile;

computing, from entities referenced in messages authored by a plurality of users of the messaging platform, a respective plurality of trending entities for each topic of the plurality of topics, including;

computing, for each topic, a respective intra-topic weight for each entity belonging to the topic, the intra-topic weight for each entity belonging to a particular topic representing a significance of (i) a recent frequency for the entity with respect to the particular topic, compared to (ii) a baseline frequency for the entity with respect to the particular topic, and

designating, for each topic, one or more entities having an intra-topic weight that satisfies a threshold as trending entities for the topic;

receiving, from a client device, a request for trending content to be provided to a user device associated with the profile of a messaging platform;

computing a profile-specific score for each trending entity of a plurality of trending entities belonging to the topics of interest determined for the profile, including multiplying a respective intra-topic weight for the trending entity by an intra-profile weight of the profile for a topic of the trending entity;

selecting, for each topic, a respective number of trending entities based on the respective profile-specific scores computed for the plurality of trending entities belonging to the topics of interest determined of the profile;

retrieving one or more messages associated with the selected trending entities; and

sending, to the client device, content associated with one or more of the retrieved messages for display to the user of the profile in a message stream, thereby reducing a likelihood of the user missing content that reflects trending entities in topics that are of interest to the user.

25. The messaging platform system of claim 24, wherein the one or more computers are further configured to perform operations comprising:

storing a plurality of whitelisted domains in a whitelist.

26. The messaging platform system of claim 25, wherein the one or more computers are configured to:

identify, by parsing a message in the one or more messages, a link in the message to a news story and a source of the news story;

compare the source of the news story against a plurality of whitelisted domains; and

add, in response to determining that the source of the news story is in the plurality of whitelisted domains, the link to the content for display to the user of the profile.

27. The system of claim 24, wherein the profile is a first profile, and the intra-profile weight for a particular topic of interest is determined based on (i) a number of messages of the messaging platform that are associated with the particular topic and are issued or forwarded by the first profile, or (ii) a number of other profiles that are followed by the first profile and are known for issuing messages associated with the particular topic of interest.

28. The system of claim 24, wherein a particular trending entity is associated with a first topic and a second topic of the plurality of topics of interest, and wherein computing a profile-specific score for the particular trending entity includes calculating a combined weight for the particular trending entity by summing (i) a first profile-specific score of the particular trending entity for the first topic, and (ii) a second profile-specific score of the particular trending entity for the second topic.

29. The system of claim 24, wherein the operations further comprise:

for each topic, calculating a list size, the list size setting a maximum number (N) of trending entities to be selected for the topic, wherein the trending entities selected for the topic comprise N top trending entities associated with a top N intra-topic weights with respect to the topic.

30. The system of claim 24, wherein a first entity belongs to a first topic and to a second topic different from the first topic, the first entity having a first intra-topic weight associated with the first topic, and a second intra-topic weight associated with the second topic.