Video recommendation based on video co-occurrence statistics

1. A computer method for generating video recommendations for a video in a video sharing environment, comprising:

detecting a target video viewed by a user;

generating video co-occurrence data of the target video, the video co-occurrence data including information associated with the target video related to one or more other videos on one or more video playlists;

selecting one or more co-occurrence videos associated with the target video based on the video co-occurrence data of the target video, wherein each selected co-occurrence video of the target video is a video on a video playlist containing the target video and a distance on the playlist between each selected co-occurrence video and the target video is determined based on an order in which the target video and the selected co-occurrence video are to be played;

ranking the selected co-occurrence videos based at least upon the distances associated with the selected co-occurrence videos; and

generating one or more video recommendations for the target video based on the ranking of the selected co-occurrence videos.

2. The method of claim 1 , wherein generating video co-occurrence data of the target video comprises:

identifying one or more video playlists containing the target video;

for each identified video playlist, generating one or more video pairs, each video pair containing the target video and another video on the video playlist; and

for each video pair, computing the frequency of the video pair on the one or more video playlists.

3. The method of claim 1 , wherein generating video co-occurrence data of the target video further comprises:

eliminating duplicate video pairs on a video playlist, a duplicate video pair containing two same videos as another video pair.

4. The method of claim 1 , wherein one or more video playlists are generated by a user of the video sharing environment.

5. The method of claim 1 , wherein one or more video playlists are generated by a network entity of the video sharing environment.

6. The method of claim 1 , wherein a video playlist is described by metadata associated with the video playlist, the metadata identifying one or more videos on the video playlist and order of the videos to be played.

7. The method of claim 1 , further comprising ranking the one or more video playlists based on the video co-occurrence data of videos contained on the video playlists.

8. A non-transitory computer-readable storage medium storing executable computer program instructions for generating video recommendations for a video in a video sharing environment, the computer program instructions comprising instructions for:

detecting a target video viewed by a user;

generating video co-occurrence data of the target video, the video co-occurrence data including information associated with the target video related to one or more other videos on one or more video playlists;

selecting one or more co-occurrence videos associated with the target video based on the video co-occurrence data of the target video, wherein each selected co-occurrence video of the target video is a video on a video playlist containing the target video and a distance on the playlist between each selected co-occurrence video and the target video is determined based on an order in which the target video and the selected co-occurrence video are to be played;

ranking the selected co-occurrence videos based at least upon the distances associated with the selected co-occurrence video; and

generating one or more video recommendations for the target video based on the ranking of the selected co-occurrence videos.

9. The computer-readable storage medium of claim 8 , wherein the computer program instructions for generating video co-occurrence data of the target video comprises instructions for:

identifying one or more video playlists containing the target video;

for each identified video playlist, generating one or more video pairs, each video pair containing the target video and another video on the video playlist; and

for each video pair, computing the frequency of the video pair on the one or more video playlists.

10. The computer-readable storage medium of claim 8 , wherein the computer program instructions for generating video co-occurrence data of the target video further comprises instructions for:

eliminating duplicate video pairs on a video playlist, a duplicate video pair containing two same videos as another video pair.

11. The computer-readable storage medium of claim 8 , wherein a video playlist is described by metadata associated with the video playlist, the metadata identifying one or more videos on the video playlist and order of the videos to be played.

12. The computer-readable storage medium of claim 8 , further comprising computer program instructions for ranking the one or more video playlists based on the video co-occurrence data of videos contained on the video playlists.

13. A system for generating video recommendations for a video in a video sharing environment, comprising:

a non-transitory computer-readable storage medium storing executable computer modules, comprising:

a video co-occurrence module for:

detecting a target video viewed by a user; and

generating video co-occurrence data of the target video, the video co-occurrence data including information associated with the target video related to one or more other videos on one or more video playlists a ranking module for:

selecting one or more co-occurrence videos associated with the target video based on the video co-occurrence data of the target video, wherein each selected co-occurrence video of the target video is a video on a video playlist containing the target video and a distance on the playlist between each selected co-occurrence video and the target video is determined based on an order in which the target video and the selected co-occurrence video are to be played; and

ranking the selected co-occurrence videos based at least upon the distances associated with the selected co-occurrence video;

a recommendation module for generating one or more video recommendations for the target video based on the ranking of the selected co-occurrence videos; and

a computer processor configured to execute the computer modules.

14. The system of claim 13 , wherein the video co-occurrence module is further for:

identifying one or more video playlists containing the target video;

for each identified video playlist, generating one or more video pairs, each video pair containing the target video and another video on the video playlist; and

for each video pair, computing the frequency of the video pair on the video playlists.

15. A method for generating video recommendations for a video in a video sharing environment, the method comprising:

detecting a target video viewed by a user;

generating video co-occurrence data for the target video, the video co-occurrence data including information associated with the target video related to one or more other videos on one or more video playlists;

selecting one or more co-occurrence videos associated with the target video based on the video co-occurrence data of the target video, wherein each selected co-occurrence video of the target video is a video on a video playlist containing the target video and a distance on the playlist between each selected co-occurrence video and the target video is determined based on an order in which the target video and the selected co-occurrence video are to be played;

ranking the selected co-occurrence videos based on a frequency with which each of the co-occurrence videos is paired with the target video on the video playlists; and

generating one or more video recommendations for the target video based on the ranking of the selected co-occurrence videos.

16. A method for generating video recommendations for a video in a video sharing environment, the method comprising:

detecting a target video viewed by a user;

generating video co-occurrence data for the target video, the video co-occurrence data including information associated with the target video related to one or more other videos on one or more video playlists;

selecting one or more co-occurrence videos associated with the target video based on the video co-occurrence data of the target video, wherein each selected co-occurrence video of the target video is a video on a video playlist containing the target video and a distance on the playlist between each selected co-occurrence video and the target video is determined based on an order in which the target video and the selected co-occurrence video are to be played;

determining an aggregate ranking score for each of the selected co-occurrence videos, the aggregate ranking score determined according to a plurality of weighted ranking factors;

ranking the selected co-occurrence videos based on the determined ranking scores; and

generating one or more video recommendations for the target video based on the ranking of the selected co-occurrence videos.

17. The method of claim 16 wherein the plurality of weighted ranking factors includes a video uploading time.

18. The method of claim 16 wherein the plurality of weighted ranking factors includes video quality.