Use of user consumption time to rank media suggestions

1. A method comprising:

presenting a list of one or more videos via a user interface;

receiving a selection of a target video to watch from the list;

playing the target video in the user interface; and

updating the user interface to present one or more suggested videos concurrently with playback of the target video, wherein the one or more suggested videos are predicted to be watched by a user for at least a threshold duration based on user consumption times of a first plurality of videos selected by a user and user consumption times of a second plurality of videos suggested to the user in association with respective videos from the first plurality of videos.

2. The method of claim 1 , further comprising:

receiving a selection of one of the one or more suggested videos as a new target video to watch;

playing the new target video in the user interface; and

updating the user interface to present one or more additional suggested videos concurrently with playback of the new target video, wherein the one or more additional suggested videos are predicted to be watched by the user for at least the threshold duration.

3. The method of claim 1 , wherein the one or more suggested videos are selected from a plurality of candidate videos that correspond to the target video.

4. The method of claim 3 , wherein-the one or more suggested videos are predicted to be watched by the user for at least the threshold duration responsive to application of a prediction model to features of the plurality of candidate videos, wherein the prediction model is configured to predict a length of time each candidate video is to be watched by the user, and the one or more suggested videos are selected as a subset of the plurality of candidate videos with predicted lengths of time that satisfy the threshold duration.

5. The method of claim 4 , wherein predicting the length of time each candidate video is to be watched by the user further comprises:

identifying, for each candidate video, other videos that are suggested in association with the corresponding candidate video, wherein the other suggested candidate videos are child candidate videos and the corresponding candidate video is a parent candidate video; and

predicting a sub-tree length of time that each candidate video is to be watched by the user, wherein the sub-tree length of time comprises a total of the length of time for the parent candidate video and the length of time of each child candidate video.

6. The method of claim 4 , wherein, for each candidate video of the plurality of candidate videos, the features comprise at least one of an identifier of the target video corresponding to the candidate video, an identifier of each suggested video of the one or more suggested videos that is in a same set of suggestions as the candidate video, a length of time the other candidate videos are watched, a country identifier, a gender of a user associated with the candidate video, or an age of the user associated with the candidate video.

7. The method of claim 4 , wherein the subset comprises a predefined number of top ranking candidate videos.

8. The method of claim 4 , wherein predicting the length of time each candidate video is to be watched by the user further comprises:

determining, for each candidate video, a percentage of the candidate video that is to be consumed by the user;

determining which candidate videos satisfy a percentage threshold; and

predicting the length of time for each candidate video that satisfies the percentage threshold.

9. The method of claim 8 , wherein identifying which candidate videos satisfy the percentage threshold further comprises:

determining, for each candidate video, an average percentage of the candidate video that is to be consumed by the user; and

identifying candidate videos comprising an average percentage that satisfies the percentage threshold.

10. A system comprising:

a memory; and

a processing device coupled to the memory to:

present a list of one or more videos via a user interface;

receive a selection of a target video to watch from the list;

play the target video in the user interface; and

update the user interface to present one or more suggested videos concurrently with playback of the target video, wherein the one or more suggested videos are predicted to be watched by a user for at least a threshold duration based on user consumption times of a first plurality of videos selected by a user and user consumption times of a second plurality of videos suggested to the user in association with respective videos from the first plurality of videos.

11. The system of claim 10 , wherein the processing device further to:

receive a selection of one of the one or more suggested videos as a new target video to watch;

play the new target video in the user interface; and

update the user interface to present one or more additional suggested videos concurrently with playback of the new target video, wherein the one or more additional suggested videos are predicted to be watched by the user for at least the threshold duration.

12. The system of claim 10 , wherein the one or more suggested videos are selected from a plurality of candidate videos that correspond to the target video.

13. The system of claim 10 , wherein the one or more suggested videos are predicted to be watched by the user for at least the threshold duration responsive to application of a prediction model to features of the plurality of candidate videos, wherein the prediction model is configured to predict a length of time each candidate video is to be watched by the user, and the one or more suggested videos are selected as a subset of the plurality of candidate videos with predicted lengths of time that satisfy the threshold duration.

14. The system of claim 13 , wherein to predict the length of time each candidate video is to be watched by the user, a server further to:

identify, for each candidate video, other videos that are suggested in association with the corresponding candidate video, wherein the other suggested candidate videos are child candidate videos and the corresponding candidate video is a parent candidate video; and

predict a sub-tree length of time that each candidate video is to be watched by the user, wherein the sub-tree length of time comprises a total of the length of time for the parent candidate video and the length of time of each child candidate video.

15. The system of claim 13 , wherein, for each candidate video of the plurality of candidate videos, the features comprise at least one of an identifier of the target video corresponding to the candidate video, an identifier of each suggested video of the one or more suggested videos that is in a same set of suggestions as the candidate video, a length of time the other candidate videos are watched, a country identifier, a gender of a user associated with the candidate video, or an age of the user associated with the candidate video.

16. The system of claim 13 , wherein the subset comprises a predefined number of top ranking candidate videos.

17. A non-transitory, computer readable medium storing instructions that, when executed by a processing device, cause the processing device to:

present a list of one or more videos via a user interface;

receive a selection of a target video to watch from the list;

play the target video in the user interface; and

update the user interface to present one or more suggested videos concurrently with playback of the target video, wherein the one or more suggested videos are predicted to be watched by a user for at least a threshold duration based on user consumption times of a first plurality of videos selected by a user and user consumption times of a second plurality of videos suggested to the user in association with respective videos from the first plurality of videos.

18. The computer readable medium of claim 17 , wherein the processing device further to:

receive a selection of one of the one or more suggested videos as a new target video to watch;

play the new target video in the user interface; and

update the user interface to present one or more additional suggested videos concurrently with playback of the new target video, wherein the one or more additional suggested videos are predicted to be watched by the user for at least the threshold duration.

19. The computer-readable medium of claim 17 , wherein the one or more suggested videos are selected from a plurality of candidate videos that correspond to the target video.

20. The computer-readable medium of claim 19 , wherein the one or more suggested videos are predicted to be watched by the user for at least the threshold duration responsive to application of a prediction model to features of the plurality of candidate videos, wherein the prediction model is configured to predict a length of time each candidate video is to be watched by the user, and the one or more suggested videos are selected as a subset of the plurality of candidate videos with predicted lengths of time that satisfy the threshold duration.