Network-based content discovery using messages of a messaging platform

Patent metadata

Field	Value
Publication number	US-9509643-B1
Application number	US-201314078239-A
Country	US
Kind code	B1
Filing date	Nov 12, 2013
Priority date	Nov 12, 2013
Publication date	Nov 29, 2016
Grant date	Nov 29, 2016

Abstract

A system, method, and computer-readable medium for identifying relevant content from a messaging platform. The method can include: identifying a context account; identifying a set of initial accounts of the messaging platform; selecting a set of relevant accounts from among the set of initial accounts; selecting a set of messages authored by the set of relevant accounts based at least on a recency of each of the set of messages; and providing the set of messages in response to a request.

First claim

What is claimed is:

1. A method for identifying relevant content from a messaging platform, comprising:
receiving a message from an account of the messaging platform;
selecting, in response to receiving the message, a context account when the context account follows the account in a connection graph;
inserting the message into a stream associated with the context account;
identifying a set of initial accounts, wherein each account in the set of initial accounts does not have a following relationship with the context account within the connection graph and wherein each of the set of initial accounts is also one selected from a group consisting of a non-spam account, a top-ranked account, and a high-credibility account;
selecting, by a computer processor, a set of relevant accounts from among the set of initial accounts by weighting the set of initial accounts based at least on a first criterion, wherein the first criterion comprises engagement of accounts in the initial set of accounts;
selecting a set of messages authored by the set of relevant accounts based at least on a second criterion, wherein the second criterion comprises engagement involving messages in the set of messages; and
inserting the set of messages into the stream associated with the context account.

2. The method of claim 1 ,
wherein selecting the set of relevant accounts from among the set of initial accounts comprises:
initializing a relevance score based on the at least the first criterion.

3. The method of claim 2 , wherein selecting the set of relevant accounts further comprises:
ranking the initial accounts based on their corresponding relevance scores; and
selecting the set of relevant accounts from the set of initial accounts based on the ranking of the initial accounts.

4. The method of claim 3 , wherein selecting the set of messages comprises:
identifying a set of candidate messages authored by the set of relevant accounts;
weighting the candidate messages based on the at least one second criterion; and
selecting the set of messages using the weighting of the candidate messages.

5. The method of claim 2 , wherein weighting the initial accounts comprises weighting the relevance score based on at least one selected from a group consisting of a geographic coordination associated with one or more initial accounts, a popularity of one or more initial accounts, a number of connections of one or more initial accounts, and a ratio of inbound graph connections to outbound graph messages.

6. The method of claim 1 , wherein the following relationship specifies that the context account is a distance of one from the account in the connection graph.

7. A system for identifying relevant content based on messages from a messaging platform, comprising:
a computer processor;
a content discovery module executing on the computer processor and configured to enable the computer processor to:
receive a message from an account of the messaging platform;
selecting, in response to receiving the message, a context account when the context account follows the account in a connection graph;

insert the message into a stream associated with the context account;
identify a set of initial accounts, wherein each account in the set of initial accounts does not have a following relationship with the context account within the connection graph and wherein each of the set of initial accounts is also one selected from a group consisting of a non-spare account, a top-ranked account, and a high-credibility account;
select a set of relevant accounts from among the set of initial accounts by weighting the set of initial accounts based at least on a first criterion, wherein the first criterion comprises engagement of accounts in the initial set of accounts;
select a set of messages authored by the set of relevant accounts based at least on a second criterion, wherein the second criterion comprises engagement involving messages in the set of messages; and
insert the set of messages into the stream associated with the context account.

8. The system of claim 7 ,
wherein selecting the set of relevant accounts from among the set of initial accounts comprises:
initializing a relevance score based on the at least the first criterion.

9. The system of claim 8 , wherein selecting the set of relevant accounts further comprises:
rank the initial accounts based on their corresponding relevance scores; and
select the set of relevant accounts from the set of initial accounts based on the ranking of the initial accounts.

10. The system of claim 9 , wherein to select the set of messages, the content discover module is further configured to enable the computer processor to:
identify a set of candidate messages authored by the set of relevant accounts;
weight the candidate messages based on the at least one second criterion; and
select the set of messages using the weighting of the candidate messages.

11. The system of claim 8 , wherein to weight the initial accounts, the content discovery module is further configured to enable the computer processor to weight the relevance score based on at least one selected from a group consisting of a geographic coordination associated with one or more initial accounts, a popularity of one or more initial accounts, a number of connections of one or more initial accounts, and a ratio of inbound graph connections to outbound graph messages.

12. The system of claim 7 , wherein the following relationship specifies that the context account is a distance of one from the account in the connection graph.

13. A non-transitory computer readable medium comprising computer readable program code for:
receiving a message from an account of a messaging platform;
selecting, in response to receiving the message, a context account when the context account follows the account in a connection graph;
inserting, based on the relationship, the message into a stream associated with the context account;
identifying a set of initial accounts wherein each account in the set of initial accounts does not have a following relationship with the context account within the connection graph and wherein each of the set of initial accounts is also one selected from a group consisting of a non-spam account, a top-ranked account, and a high-credibility account;
selecting, by a computer processor, a set of relevant accounts from among the set of initial accounts by weighting the set of initial accounts based at least on a first criterion, wherein the first criterion comprises engagement of accounts in the initial set of accounts;
selecting a set of messages authored by the set of relevant accounts based at least on a second criterion, wherein the second criterion comprises engagement involving messages in the set of messages; and
inserting providing the set of messages into the stream associated with the context account.

14. The non-transitory computer readable medium of claim 13 , wherein selecting the set of relevant accounts from among the set of initial accounts comprises:
initializing a relevance score based on the at least the first criterion.

15. The non-transitory computer readable medium of claim 14 , wherein selecting the set of messages further comprises computer readable program code for:
identifying a set of candidate messages authored by the set of relevant accounts;
weighting the candidate messages based on the at least one second criterion; and
selecting the set of messages using the weighting of the candidate messages.

16. The non-transitory computer readable medium of claim 14

Assignees

• Twitter Inc

Inventors

• Gade Krishna
• Loreto Daniel
• Callari Francesco G
• Marinus Julian M
• Li Yu-Shuang

Classifications

• H04L51/00 Primary

  User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail · CPC title

• H04L51/52

  for supporting social networking services · CPC title

• H04L51/214 Primary

  using selective forwarding · CPC title

• H04L67/535

  Tracking the activity of the user (network monitoring arrangements H04L43/00; recording of computer activity G06F11/34) · CPC title

• H04L67/306

  User profiles · CPC title