Visual presentation of video recommendations

• US 7,966,632 B1
• Filed: 12/12/2007
• Issued: 06/21/2011
• Est. Priority Date: 12/12/2007
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Pan, Yu, Schrock, Jasson
• Primary Examiner(s): Beliveau; Scott
• Assistant Examiner(s): Stanley; Mark P
• Application Number:
  US11/955,298
• Time in Patent Office: 1,287 Days
• Field of Search: 725/46
• US Class Current: 725/46
• CPC Class Codes: H04N 21/252 Processing of multiple end-users' preferences to derive collaborative data
• CPC Class Codes: H04N 21/25891 being end-user preferences retrieval of video data in a video database based on user preferences G06F16/739; arrangements for recognizing users' preferences H04H60/46; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18
• CPC Class Codes: H04N 21/4312 involving specific graphical features, e.g. screen layout, special fonts or colors, blinking icons, highlights or animations
• CPC Class Codes: H04N 21/47 End-user applications
• CPC Class Codes: H04N 21/4826 using recommendation lists, e.g. of programs or channels sorted out according to their score

Abstract

A system, method and various user interfaces enable visually browsing multiple groups of video recommendations. A group of video recommendations includes a central video and a set of related videos displayed at its corresponding display distance from the central video and from each other. The corresponding display distance of a video is calculated as a function of the video'"'"'s recommendation score and correlation with other videos of the group. The recommendation score is generated based on a weighted covisitation measure. Visually presenting recommended videos to a user provides a quick and intuitive way to find videos of interest and visualize the browsing of the recommended videos.

Claims

1. A computer-implemented method for visually browsing a plurality of groups of video recommendations, comprising:

receiving a selection of a video from among a first group of video recommendations;

designating the selected video as a central video;

generating a second group of video recommendations in response to the selection of the central video, each video in the second group of video recommendations associated with a recommendation score; and

coordinating a visual display of the second group of video recommendations so that videos in the second group are displayed around the central video in a display area, wherein each video in the second group of videos is displayed at a distance from the central video based on its associated recommendation score relative to the central video, and displayed in the display area relative to each of the other videos in the second group based upon its recommendation score relative to the recommendation score of each of the other videos in the second group of videos.

2. The method of claim 1, further comprising visually displaying the first group of video recommendations in the display area.

3. The method of claim 1, wherein the second group of video recommendations comprises a set of videos, at least one video of the set having a pulling and repelling force towards at least one other video of the set.

4. The method of claim 1, wherein generating the second group of video recommendation further comprises:

relating a video of the second group to another video of the second group based on a correlation between the videos, the correlation being determined by pulling and repelling forces between the videos.

5. The method of claim 4, wherein a pulling force of a first video towards a second video measures the relative relatedness between the videos, the pulling force of a video being proportional to the recommendation score of the video.

6. The method of claim 4, wherein a repelling force of a first video towards a second video measures the relative unrelatedness between the videos, the repelling force of a video being proportional to the recommendation score of the video.

7. The method of claim 1, wherein generating the second group of video recommendations comprises:

for a plurality of videos, relating the video to the central video using a weighted covisitation measure; and

generating a recommendation score for the video based on the weighted covisitation measure.

8. The method of claim 7, wherein generating a recommendation score of a video comprises calculating a probability of the video being viewed next by a client conditioned on the central video being viewed.

9. The method of claim 8, wherein the probability of the video being viewed next conditioned on the central video being viewed is

10. The method of claim 7, wherein the recommendation score of the video is generated at least in part based on length of the video actually watched.

11. The method of claim 7, wherein the recommendation score of the video is generated at least in part based on length of the video being recommended.

12. The method of claim 7, wherein the recommendation score of the video is generated at least in part based on difference in viewing time between the video being recommended and the related central video.

13. The method of claim 1, wherein coordinating the visual display of second group of video recommendations comprises:

adjusting the respective distance of a video of the second group of videos relating to a plurality of central videos in the display area; and

removing one or more video recommendations from the display area in response to a total number of video recommendations to be displayed exceeding a threshold value.

14. The method of claim 13, wherein adjusting the respective distance of a video related to a plurality of central videos comprises recalculating the recommendation score of the video based on a weighted covisitation measure of the video and a plurality of central videos being related to the video.

15. A computer system for visually browsing a plurality of groups of video recommendations, comprising:

a computer-readable storage medium storing executable computer modules, comprising;

a receiving module configured to;

receive from a client computer a selection of a video from a first group of video recommendations; and

designate the selected video as a central video;

a recommendation module configured to generate a second group of video recommendations in response to the user selection of the central video, each video in the second group of video recommendations associated with a recommendation score;

a presentation module configured to coordinate visual display of the second group of video recommendations in a display area so that videos in the second group of videos are displayed around the selected central video in the display area, wherein each video in the second group of videos is displayed at a distance from the central video based on its associated recommendation score relative to the central video, and displayed in the display area relative to each other videos in the second group based upon its recommendation score relative to the recommendation score of each of the other videos in the second group of videos; and

a processor configured to execute the computer modules stored by the computer-readable storage medium.

16. The system of claim 15, further comprises a display module configured to visually display the first group of video recommendations in the display area.

17. The system of claim 15, wherein the second group of video recommendations comprises a set of videos, at least one video of the set having a pulling and repelling force towards at least one other video of the set.

18. The system of claim 15, wherein the recommendation module is further configured to:

relating a video of the second group to another video of the second group based on a correlation between the videos, the correlation being determined by the pulling and repelling forces between the videos.

19. The system of claim 18, wherein a pulling force and a repelling force of a video towards another video measure the correlation between the videos, the pulling and repelling force of a video being proportional to the recommendation score of the video.

20. The system of claim 15, wherein the recommendation module is configured to:

relate a video of the set to the central video using a weighted covisitation measure, a covisitation measure of a video representing a number of viewers that have viewed the video during a time period, the covisitation measure of the video being weighted by a scaling factor; and

generate a recommendation score based on the weighted covisitation measure.

21. The system of claim 20, wherein the recommendation module is configured to calculate a probability of the video being viewed next by a client set conditioned on the related central video being viewed.

22. The system of claim 20, wherein the recommendation module is further configured to generate a recommendation score of a video based on at least one of the following factors:

length of the video actually watched;

length of the video being recommended; and

difference in viewing time between the video being recommended and the related central video.

23. The system of claim 15, wherein a presentation module is configured to:

adjust the respective distance of a video of the second group of videos relating to a plurality of central videos in the display area; and

remove one or more video recommendations from the display area in response to total number of video recommendations to be displayed exceeding a threshold value.

24. The system of claim 23, wherein the presentation module is further configured to recalculate the recommendation score of the video based on a weighted covisitation measure of the video and a plurality of central videos being related to the video.

25. A non-transitory computer-readable storage medium containing a computer program product for visually browsing a plurality of groups of video recommendations, the computer program product comprising:

computer program code for visually displaying a first group of video recommendations in a display area;

computer program code for designating the selected video as a central video;

computer program code for receiving from a client computer a selection of a central video from the first group of video recommendations;

computer program code for generating a second group of video recommendations in response to the user selection of the central video, each video in the second group of video recommendation associated with a recommendation score; and

computer program code for coordinating visual display of the second group of video recommendations so that the videos in the second group of videos are displayed around the selected central video in the display area, wherein each video in the second group of videos is displayed at a distance from the central video based on its associated recommendation score relative to the central video, and displayed in the display area relative to each other videos in the second group based upon its recommendation score relative to the recommendation score of each of the other videos in the second group of videos.

26. A method for relating a plurality of video recommendations through probabilistic analysis, comprising:

receiving a selection of a first video recommendation from the plurality of video recommendations;

evaluating occurrences of the first video recommendation being viewed by a plurality of viewers;

receiving a selection of a second video recommendation from the plurality of video recommendations;

evaluating occurrences of the second video recommendation being viewed by the plurality of viewers set conditioned on the first video recommendation being viewed by the same plurality of viewers first;

calculating a probability for the second video recommendation based on the occurrences of the first video recommendation and the occurrences of the second video recommendation being viewed by the plurality of viewers; and

displaying the first video recommendation in a display area

displaying the second video recommendation in the display area at a distance from the first video recommendation based on the probability associated with the second video recommendation relative to the first video recommendation.

27. The method of claim 26, further comprising:

receiving a selection of a third video recommendation from the plurality of video recommendations;

evaluating occurrences of the third video recommendation being viewed by the plurality of viewers set conditioned on the first video recommendation being viewed by the same plurality of viewers first; and

calculating a probability for the third video recommendation based on the occurrences of the first video recommendation and the occurrences of the third video recommendation being viewed by the plurality of viewers.

28. The method of claim 26, wherein evaluating occurrences of the first video recommendation comprising using a scaling factor, wherein the scaling factor is a configurable threshold.

29. The method of claim 28, wherein the scaling factor is 0.6.

30. The method of claim 26, further comprising displaying the third video recommendation in the display area at a distance from the second video recommendation based on the probability associated with the third video recommendation relative to the second video recommendation.

31. A computer system for relating a plurality of video recommendations through probabilistic analysis, comprising:

a computer-readable storage medium storing executable computer modules, comprising;

a selection module, configured to;

receive a selection of a first video recommendation from the plurality of video recommendations; and

receive a selection of a second video recommendation from the plurality of video recommendation;

an evaluation module, configured to;

evaluate occurrences of the first video recommendation being viewed by a plurality of viewers;

evaluate occurrences of the second video recommendation being viewed by the plurality of viewers set conditioned on the first video recommendation being viewed by the same plurality of viewers first, the evaluation module being communicatively coupled to the selection module; and

a recommendation engine, communicatively coupled to the evaluation module, configured to;

calculate a probability for the second video recommendation based on the occurrences of the first video recommendation and the occurrences of the second video recommendation being viewed by the plurality of viewers;

display the first video recommendation in a display area;

display the second video recommendation in a display area at a distance from the first video recommendation based on the probability associated with the second video recommendation relative to the first video recommendation; and

a processor configured to execute the computer modules stored by the computer-readable storage medium.

32. The system of claim 31, wherein the recommendation engine is further configured to:

receive a selection of a third video recommendation from the plurality of video recommendations;

evaluate occurrences of the third video recommendation being viewed by the plurality of viewers set conditioned on the first video recommendation being viewed by the same plurality of viewers first; and

calculate a probability for the third video recommendation based on the occurrences of the first video recommendation and the occurrences of the third video recommendation being viewed by the plurality of viewers.

33. The system of claim 31, wherein the evaluation module is further configured to use a scaling factor to evaluate the occurrences of the first video recommendation being viewed by the plurality of viewers, wherein the scaling factor is a configurable threshold.

34. The system of claim 31, wherein the recommendation engine is further configured to display the third video recommendation in the display area at a distance from the second video recommendation based on the probability associated with the third video recommendation relative to the second video recommendation.

35. A non-transitory computer-readable storage medium containing a computer program product for relating a plurality of video recommendations through probabilistic analysis, the computer program product comprising:

computer program code for receiving a selection of a first video recommendation from the plurality of video recommendations;

computer program code for evaluating occurrences of the first video recommendation being viewed by a plurality of viewers;

computer program code for receiving a selection of a second video recommendation from the plurality of video recommendations;

computer code for evaluating occurrences of the second video recommendation being viewed by the plurality of viewers set conditioned on the first video being viewed by the same plurality of viewers first;

computer code for calculating a probability for the second video recommendation based on the occurrences of the first video and the occurrences of the second video recommendation being viewed by the plurality of viewers;

computer program code for displaying the first video recommendation in a display area; and

computer code for displaying the second video recommendation in the display area at a distance from the first video recommendation based on the probability associated with the second video recommendation relative to the first video recommendation.

36. The computer program product of claim 35, further comprising computer program code for:

receiving a selection of a third video recommendation from the plurality of video recommendations;

evaluating occurrences of the third video recommendation being viewed by the plurality of viewers set conditioned on the first video recommendation being viewed by the same plurality of viewers first; and

calculating a probability for the third video recommendation based on the occurrences of the first video recommendation and the occurrences of the third video recommendation being viewed by the plurality of viewers.

37. The computer program product of claim 35, further comprising computer program code for display the third video recommendation in the display area at a distance from the second video recommendation based on the probability associated with the third video recommendation relative to the second video recommendation.