Recommendations based on topic clusters

• US 8,589,434 B2
• Filed: 12/01/2011
• Issued: 11/19/2013
• Est. Priority Date: 12/01/2010
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Liebald, Benjamin, Nandy, Palash, Sampath, Dasarathi, Liu, Junning, Niu, Ye, Ilvento, Christina, Chen, Yu-To
• Primary Examiner(s): DANG, THANH HA T
• Application Number:
  US13/309,481
• Publication Number:
  US 20120143911A1
• Time in Patent Office: 719 Days
• Field of Search: 707/733, 707/734, 707/736, 707/737, 707/771, 707/708
• US Class Current: 707/771
• CPC Class Codes: G06F 15/16 Combinations of two or more digital computers each having at least an arithmetic unit, a program unit and a register, e.g. for a simultaneous processing of several programs coordinating program control therefor G06F9/52; in regulating and control system G05B
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/337 Profile generation, learning or modification
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/635 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/735 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/7867 using information manually generated, e.g. tags, keywords, comments, title and artist information, manually generated time, location and usage information, user ratings
• CPC Class Codes: G06F 3/0482 Interaction with lists of selectable items, e.g. menus
• CPC Class Codes: H04L 65/60 Network streaming of media packets
• CPC Class Codes: H04L 65/80 Responding to QoS

Abstract

A system and method for developing a user'"'"'s profile based on the user'"'"'s interaction with content items. A module on the client rendering the content items or the service including the content items tracks the user'"'"'s interactions with the content items and transmits the tracked data to a user analysis module. The user analysis module determines the topics associated with the interacted upon content items. The user analysis module then selects the topics for the user'"'"'s profiles based on the received tracked data and the associated topics. The selected topics are mapped to topic clusters and the topic clusters are stored in association with the user profile. Recommendations for a user are made based on the topic clusters associated with the user'"'"'profile.

Claims

1. A computer-implemented method for developing a profile of a user including topic clusters, the method comprising:

storing, by a computer, a plurality of topic clusters, the plurality of topic clusters including a set of related topics;

retrieving a profile of a user, the profile including a plurality of topics indicating interests of the user;

identifying, from the plurality of topic clusters, a topic cluster including a topic from the plurality of topics included in the profile of the user;

adding the identified topic cluster to a list of topic clusters for the profile of the user;

identifying a plurality of users, each of the plurality of users associated with a profile that includes the identified topic cluster added to the list of topic clusters for the profile of the user;

identifying a plurality of videos interacted with by the plurality of users;

selecting a video from the plurality of videos to recommend to the user based on interactions of the plurality of users with the plurality of videos; and

recommending the selected video to the user.

2. The method of claim 1, further comprising:

generating the plurality of topic clusters based on a co-occurrence matrix, the co-occurrence matrix storing co-occurrence strengths for pairs of topics.

3. The method of claim 2, wherein generating one of the plurality of topic clusters comprises identifying a cell in the co-occurrence matrix with a highest co-occurrence strength as a first topic in the one of plurality of topic clusters and adding topics associated with the identified cell to the one of the plurality of topic clusters.

4. The method of claim 1, wherein identifying the topic cluster comprises identifying the topic cluster based on topic association strengths of the topics in the topic cluster.

5. The method of claim 4, wherein a topic association strength for a topic in the topic cluster indicates a degree of association between the user and the topic.

6. The method of claim 1, wherein the video is selected based on a number of the plurality of users who have watched the video exceeding a threshold.

7. The method of claim 1, further comprising:

selecting a topic for recommendation to the user based on an aggregated interaction measure for videos associated with the topic by the plurality of users.

8. The method of claim 1, wherein the profile of the user includes a user cluster strength for each of the topic clusters in the list of topic clusters for the profile of the user, the user cluster strength indicating the degree of association of the cluster with the user.

9. A computer system for developing a profile of a user including topic clusters, comprising:

a non-transitory computer-readable storage medium storing executable computer program instructions comprising instructions for;

storing a plurality of topic clusters, the plurality of topic clusters including a set of related topics;

retrieving a profile of a user, the profile including a plurality of topics indicating interests of the user;

identifying, from the plurality of topic clusters, a topic cluster including a topic from the plurality of topics included in the profile of the user;

adding the identified topic cluster to a list of topic clusters for the profile of the user;

identifying a plurality of users, each of the plurality of users associated with a profile that includes the identified topic cluster added to the list of topic clusters for the profile of the user;

identifying a plurality of videos interacted with by the plurality of users;

selecting a video from the plurality of videos to recommend to the user based on interactions of the plurality of users with the plurality of videos; and

recommending the selected video to the user;

and

a processor for executing the computer program instructions.

10. The system of claim 9 further comprising:

generating the plurality of topic clusters based on a co-occurrence matrix, the co-occurrence matrix storing co-occurrence strengths for pairs of topics.

11. The system of claim 10, wherein generating one of the plurality of topic clusters comprises identifying a cell in the co-occurrence matrix with a highest co-occurrence strength as a first topic in the one of plurality of topic clusters and adding topics associated with the identified cell to the one of the plurality of topic clusters.

12. The system of claim 9, wherein identifying 11 topic cluster comprises identifying the topic cluster based on topic association strengths of the topics in the topic cluster.

13. The system of claim 12, wherein topic association strength for a topic in the topic cluster indicates a degree of association between the user and the topic.

14. The system of claim 9, wherein the video is selected based on a number of the plurality of users that have watched the video exceeding a threshold.

15. The system of claim 9, further comprising instructions for:

selecting a topic for recommendation to the user based on an aggregated interaction measure for videos associated with the topic by the plurality of users.

16. The system of claim 9, wherein the profile of the user includes a user cluster strength for each of the topic clusters in the list of topic clusters for the profile of the user, the user cluster strength indicating the degree of association of the cluster with the user.