Engaging users by personalized composing-content recommendation

Patent metadata

Field	Value
Publication number	US-11688159-B2
Application number	US-202117224487-A
Country	US
Kind code	B2
Filing date	Apr 7, 2021
Priority date	Apr 20, 2018
Publication date	Jun 27, 2023
Grant date	Jun 27, 2023

Abstract

In one embodiment, a method includes receiving an indication of a trigger action by a first user at a client system, wherein the trigger action is associated with a priming content object, identifying related content objects associated with the priming content object, selecting recommended content objects based on the priming content object, the related content objects, and profile information of the first user, wherein each of the selected recommended content objects comprises entity information of entities associated with the priming content object, and presenting content suggestions at the client system, wherein each content suggestion comprises one of the selected recommended content objects.

First claim

What is claimed is:

1. A method comprising, by a client system:
receiving, at the client system, an indication of a trigger action by a first user, wherein the trigger action is associated with a priming content object associated with one or more entities;
identifying one or more related content objects associated with the priming content object;
determining one or more recommended content objects based on an analysis by a machine-learning model of (1) the priming content object, (2) the one or more related content objects, and (3) profile information of the first user, wherein each of the recommended content objects comprises entity information of the one or more entities associated with the priming content object; and
presenting, at the client system, one or more content suggestions, wherein each content suggestion comprises one of the recommended content objects.

2. The method of claim 1 , wherein the priming content object comprises one or more of:
a post;
a message;
an entity;
a location;
a user associated with the online social network;
a trending topic; or
a news story.

3. The method of claim 1 , wherein each related content object comprises one or more of:
a comment;
an audio clip;
an image;
a video;
a web page;
a message; or
a review.

4. The method of claim 1 , wherein the trigger action comprises one or more of:
clicking on a composer interface;
browsing a content object for more than a threshold amount of time;
uploading a content object;
updating a user status;
checking in an entity associated with the online social network; or
receiving a content object from a second user associated with the online social network.

5. The method of claim 1 , wherein each content suggestion is operable to allow the first user to select a corresponding recommended content object.

6. The method of claim 5 , further comprising:
receiving, at the client system, a selection of a recommended content object from the first user; and
associating the recommended content object with the priming content object.

7. The method of claim 6 , wherein the priming content object comprises a post and the recommended content object comprises a comment, and wherein associating the recommended content object with the priming content object comprises inserting the comment below the post.

8. The method of claim 6 , wherein the priming content object comprises a message and the recommended content object comprises a replying message, and wherein associating the recommended content object with the priming content object comprises inserting the replying message below the message.

9. The method of claim 1 , wherein the one or more related content objects comprise one or more of:
content objects associated with an online social network;
content objects stored in a third-party system; or
content objects stored in the client system.

10. The method of claim 1 , further comprising:
generating, using the machine-learning model, a first feature vector representing the priming content object, the one or more related content objects, and profile information of the first user; and
accessing a plurality of second feature vectors representing a plurality of recommended content objects, respectively;
wherein determining the one or more recommended content objects is further based on comparisons between the first feature vector and the respective second feature vectors representing the recommended content objects.

11. The method of claim 10 , wherein the machine-learning model is based on a deep convolutional neural network, and wherein the machine-learning model is trained using a plurality of training samples, wherein each training sample comprises: (1) an input content object associated with an initial user, (2) a responding content object associated with a secondary user, wherein the responding content object corresponds to the input content object, (3) profile information of the initial user, and (4) profile information of the secondary user.

12. The method of claim 11 , wherein the input content object comprises one or more of:
a news feed post;
a news feed comment; or
a message in a messaging interface.

13. The method of claim 11 , wherein the responding content object comprises one or more of:
a news feed comment;
a news feed reply; or
a replying message in a messaging interface.

14. The method of claim 10 , wherein each of the comparisons between the first feature vector and the respective second feature vectors is based on calculating a similarity score between the first feature vector and each of the respective second feature vectors.

15. The method of claim 14 , wherein the similarity score is based on cosine similarity.

16. The method of claim 14 , further comprising ranking the second feature vectors based on the similarity scores.

17. The method of claim 16 , wherein determining selecting the one or more recommended content objects is based on the rankings of the respective second feature vectors.

18. The method of claim 1 , wherein the priming content object is authored by a second user, and wherein the first user and the second user have a pre-determined relationship within a specified degree of separation on an online social network.

19. One or more computer-readable non-transitory storage media embodying software that is operable when executed to:
receive, at the client system, an indication of a trigger action by a first user, wherein the trigger action is associated with a priming content object associated with one or more entities;
identify one or more related content objects associated with the priming content object;
determine one or more recommended content objects based on an analysis by a machine-learning model of (1) the priming content object, (2) the one or more related content objects, and (3) profile information of the first user, wherein each of the recommended content objects comprises entity information of the one or more entities associated with the priming content object; and
present, at the client system, one or more content suggestions, wherein each content suggestion comprises one of the recommended content objects.

20. A system comprising: one or more processors; and a non-transitory memory coupled to the processors comprising instructions executable by the processors, the processors operable when executing the instructions to:
receive, at the client system, an indication of a trigger action by a first user, wherein the trigger action is associated with a priming content object associated with one or more entities;
identify one or more related content objects associated with the priming content object;
determine one or more recommended content objects based on an analysis by a machine-learning model of (1) the priming content object, (2) the one or more related content objects, and (3) profile information of the first user, wherein each of the recommended content objects comprises entity information of the one or more entities associated with the priming content object; and
present, at the client system, one or more content suggestions, wherein each content suggestion comprises one of the recommended content objects.

Assignees

• Meta Platforms Inc

Inventors

• Peng Fuchun
• Han Kun
• Yang Wenhai
• Zhang Cheng
• Narayanan Vivek
• Koukoumidis Emmanouil

Classifications

• G06Q10/40

  Business processes related to social networking or social networking services · CPC title

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06N3/0464

  Convolutional networks [CNN, ConvNet] · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06Q10/42

  Determination of affinities or common interests between users · CPC title