Ranking of news feed in a mobile device based on local signals

• US 10,643,148 B2
• Filed: 06/02/2016
• Issued: 05/05/2020
• Est. Priority Date: 06/02/2016
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Marra, Christopher John, Sourov, Alexander A., Petrescu, Alexandru, Ahmed, Syed Shahbaz, Backstrom, Lars Seren
• Primary Examiner(s): Chen, Alan
• Assistant Examiner(s): Hoover, Brent Johnston
• Application Number:
  US15/171,122
• Publication Number:
  US 20170351675A1
• Time in Patent Office: 1,433 Days
• CPC Class Codes: G06F 16/9038 Presentation of query results
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 3/14 Digital output to display device ; Cooperation and interconnection of the display device with other functional units
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G09G 2350/00 Solving problems of bandwidth in display systems
• CPC Class Codes: G09G 2370/022 Centralised management of display operation, e.g. in a server instead of locally
• CPC Class Codes: G09G 2370/20 Details of the management of multiple sources of image data

Abstract

A mobile device receives ranked news feed stories from a server of a social networking system. The mobile device determines whether to adjust the ranking based on local information generated by applications executing on the mobile device or hardware of the mobile device. Examples of local information of the mobile device include information describing the mobile devices'"'"' network connection, geographical location, sensor readings, downloaded data on memory storage, and interactions of software applications executing on the mobile device. For example, if a network co-processor of a mobile device detects a slow network connection, the mobile device lowers the ranking of news feed stories expected to require further use of network bandwidth.

Claims

1. A computer-implemented method comprising:

receiving, by a local device from a social networking system, a plurality of news feed stories and an initial ranking score for each of the news feed stories that is determined by the social networking system;

retrieving, by the local device, local information associated with the local device, the local information generated by local hardware associated with the local device;

determining, by the local device, whether to modify the initial ranking score of at least one story of the plurality of news feed stories based on the retrieved local information and an attribute associated with the at least one story;

modifying, by the local device, the initial ranking score of the at least one story based on the determination;

ranking, by the local device, the plurality of news feed stories based on the modified ranking score of the at least one story and the initial ranking scores of the other news feed stories; and

displaying, on the local device, the plurality of news feed stories in an order determined based on the ranking.

2. The method of claim 1, wherein receiving, by the local device from the social networking system, a plurality of news feed stories comprises:

receiving one or more news feed stories from a server of the social networking system;

receiving one or more news feed stories stored in a hardware memory storage of the local device; and

combining the one or more news feed stories obtained from the server of the social networking system with the one or more news feed stored in the hardware memory storage of the local device to obtain the plurality of news feed stories.

3. The method of claim 1, wherein the attribute associated with the at least one story represents a measure of size of data of the at least one story.

4. The method of claim 1, wherein the attribute associated with the at least one story describes a type of multimedia included in the at least one story.

5. The method of claim 1, further comprising training a model configured to modify initial ranking scores of news feed stories, the training utilizing training data sets obtained from the local device, wherein features used for the model comprise information generated by the local hardware associated with the local device, wherein determining whether to modify the initial ranking score of the at least one story of the plurality of news feed stories is further based on the trained model.

6. The method of claim 5, wherein the model is periodically retrained at a rate determined based on interactions performed by a user on the local device.

7. The method of claim 1, wherein the local hardware is a first local hardware, the method further comprising:

retrieving, by the local device, additional local information generated by a second local hardware associated with the local device,

wherein determining whether to modify the initial ranking score of the at least one story of the plurality of news feed stories is further based on a weighted average of values associated with the local information generated by the first local hardware and the additional local information generated by the second local hardware.

8. The method of claim 1, wherein retrieving, by the local device, the local information associated with the local device comprises retrieving, from a global positioning system (GPS) coprocessor, information describing a geographical location of the local device, wherein modifying the initial ranking score of the at least one story comprises increasing the initial ranking score of the at least one story responsive to determining that the at least one story is associated with the geographical location of the local device.

9. The method of claim 1, wherein retrieving, by the local device, the local information associated with the local device comprises retrieving, from a memory storage hardware of the local device, information describing data downloaded to the memory storage hardware, wherein the data downloaded to the memory storage hardware is a portion of a content item associated with the at least one story.

10. The method of claim 9, wherein modifying the initial ranking score of the at least one story comprises increasing the initial ranking score of the at least one story responsive to determining that a download of the content item associated with the at least one story is completed.

11. The method of claim 1, wherein retrieving, by the local device, the local information associated with the local device comprises retrieving, from a hardware sensor, information describing a sensor reading of the local device.

12. The method of claim 11, wherein the hardware sensor is an ambient light sensor and the sensor reading describes the ambient lighting conditions surrounding the local device, and wherein determining whether to modify the initial ranking score of the at least one story of the plurality of news feed stories is further based on the ambient lighting conditions.

13. The method of claim 11, wherein the hardware sensor is an accelerometer and the sensor reading describes one or more axis of acceleration of the local device, and wherein determining whether to modify the initial ranking score of the at least one story of the plurality of news feed stories is further based on the one or more axis of acceleration.

14. The method of claim 11, wherein the hardware sensor is a gyroscope and the sensor reading describes one or more axis of rotation of the local device, and wherein determining whether to modify the initial ranking score of the at least one story of the plurality of news feed stories is further based on the one or more axis of rotation.

15. The method of claim 11, wherein the hardware sensor is a microphone and the sensor reading describes audio data recorded by the local device, and wherein determining whether to modify the initial ranking score of the at least one story of the plurality of news feed stories is further based on the audio data.

16. The method of claim 11, wherein the hardware sensor is a camera and the sensor reading describes a photo taken by the local device, and wherein determining whether to modify the initial ranking score of the at least one story of the plurality of news feed stories is further based on the photo.

17. A computer-implemented method comprising:

receiving, by a local device from a social networking system, a plurality of news feed stories and an initial ranking score for each of the news feed stories that is determined by the social networking system;

retrieving, by the local device, local information generated by a third party software application running on the local device, the third party software application located outside the domain of the social networking system;

determining, by the local device, whether to modify the initial ranking score of at least one story of the plurality of news feed stories based on the retrieved local information and an attribute associated with the at least one story;

modifying, by the local device, the initial ranking score of the at least one story based on the determination by inputting into a trained model features that comprise information generated by the third party software application running on the local device, the model trained based on training data sets obtained from the local device;

ranking, by the local device, the plurality of news feed stories based on the modified ranking score of the at least one story and the initial ranking scores of the other news feed stories; and

displaying, on the local device, the plurality of news feed stories in an order determined based on the ranking.

18. The method of claim 17, wherein determining whether to modify the initial ranking score of at least one story of the plurality of news feed stories is further based on the trained model.

19. The method of claim 17, wherein the model is periodically retrained at a rate determined based on interactions performed by the user on the local device.