Techniques for populating a content stream on a mobile device
View Patent ↗Techniques for populating a content stream of a mobile application are described. Consistent with some embodiments, the delays associated with waiting for a server to respond to a user's manual request for content items are eliminated by using various sensors of a mobile computing device to predict when a user is likely to view a personalized content stream. Accordingly, a background process identifies when certain conditions are satisfied, and an automatic content request is communicated to a content server, such that content items for presentation to a member of a social networking service are pre-fetched.
1. A method comprising:
at a mobile computing device, receiving from an activity recognition module, an activity status identifier derived by the activity recognition module processing information generated by one or more sensors of the mobile computing device, the activity status identifier identifying one of a plurality of possible physical activity states of a user of the mobile computing device; and
based in part on determining that a predetermined pattern of changes in the activity status identifier has occurred, communicating a content request to a server, the content request for content items to be presented in a content feed of a mobile application executing at the mobile computing device, wherein the predetermined pattern of changes in the activity status identifier has been historically observed to have a high co-occurrence with content requests for content items to be presented in the content feed of the mobile application.
2. The method of claim 1 wherein the plurality of physical activity states of the user that can be identified by an activity status identifier comprise: walking, running, sitting, standing, driving in a vehicle, and riding in a vehicle.
3. The method of claim 1 , wherein the one or more sensors comprise: a location sensor, an accelerometer, a gyroscope, a compass, a radio transceiver, and a clock.
4. The method of claim 1 , wherein determining that the predetermined pattern of changes in the ac status identifier has occurred includes determining that the received activity status identifier identifies a particular physical activity state that, based on historically observed data, frequently occurs at a location proximate to the current location of the mobile computing device and when a user initiates a content request for content items to be presented in the content feed of the mobile application.
5. The method of claim 1 , wherein determining that the predetermined pattern of changes in the activity status identifier has occurred includes determining that the received activity status identifier identifies a particular physical activity state that, based on historically observed data, frequently occurs at a time of day temporally proximate to the current time of day and when a user initiates a content request for content items to be presented in the content feed of the mobile application.
6. The method of claim 1 , wherein the predetermined pattern of changes in the activity status identifier includes an indication that the user has gone from walking to sitting still.
7. The method of claim 1 , wherein the predetermined pattern of changes in the activity status identifier includes an indication that the user has gone from driving to sitting still.
8. A method comprising:
at a mobile computing device, receiving from an activity recognition module, an activity status identifier and a confidence level, the activity status identifier and the confidence level derived by the activity recognition module processing information generated by one or more sensors of the mobile computing device, the activity status identifier identifying one of a plurality of possible physical activity states of a user of the mobile computing device and the confidence level representing the likelihood that the user is in the physical activity state identified by the activity status identifier; and
based in part on determining that a predetermined pattern of changes in the activity status identifier has occurred, communicating a content request to a server, the content request for content items to be presented in a content feed of a mobile application executing at the mobile computing device, wherein the predetermined pattern of changes in the activity status identifier has been historically observed to have a high co-occurrence with content requests for content items to be presented in the content feed of the mobile application.
9. The method of claim 8 , wherein the plurality of physical activity states of the user that can be identified by an activity status identifier comprise: walking, running, sitting, standing, driving in a vehicle, and riding in a vehicle.
10. The method of claim 8 , wherein the one or more sensors comprise: a location sensor, an accelerometer, a gyroscope, a compass, a radio transceiver, and a clock.
11. The method of claim 8 , wherein determining that the predetermined pattern of changes in the activity status identifier has occurred includes determining that the confidence level exceeds some threshold and the received activity status identifier identifies a particular physical activity state that, based on historically observed data, is associated with a high occurrence of previous content requests for content items to be presented in the content feed of the mobile application, wherein the previous content requests have occurred in a location proximate to the current location of the computing device.
12. The method of claim 8 , wherein determining that the the predetermined pattern of changes in the activity status identifier has occurred includes determining that the confidence level exceeds some threshold and the received activity status identifier identities a particular physical activity state that, based on historically observed data, is associated with a high occurrence of previous content requests for content items to be presented in the content feed of the mobile application, wherein the previous content requests have occurred at a time of day temporally proximate to the current time of day.