Methods, systems, and media for recommending media content

Patent metadata

Field	Value
Publication number	US-9424320-B2
Application number	US-201514694876-A
Country	US
Kind code	B2
Filing date	Apr 23, 2015
Priority date	Dec 31, 2012
Publication date	Aug 23, 2016
Grant date	Aug 23, 2016

Abstract

Mechanisms are provided that: identify topics associated with a plurality of pieces of media content presented in a session; calculate a distance metric for pairs of topics, wherein each of the pairs of topics includes a first topic associated with a first piece of media content and a second topic associated with a second piece of media content, and wherein the second piece of media content was presented within a given span of the presentation of the first piece of media content; for each first topic of the pairs of topics, generate a rank-ordered list for all corresponding second topics; for each of the plurality of pieces of media content, generate a single rank-ordered list of all second topics; and for each of the plurality of pieces of media content, identify one or more other pieces of media content as recommended media content based on the single rank-ordered list.

First claim

What is claimed is:

1. A system for recommending media content, comprising:
at least one hardware processor configured to:
identify topics associated with a plurality of pieces of media content presented in a session;
calculate a distance metric for pairs of topics, wherein each of the pairs of topics includes a first topic associated with a first piece of media content and a second topic associated with a second piece of media content, and wherein the second piece of media content was presented within a given span of the presentation of the first piece of media content;
for each of the plurality of pieces of media content, generate a single rank-ordered list of all second topics; and
for each of the plurality of pieces of media content, identify one or more other pieces of media content as recommended media content based on the single rank-ordered list.

2. The system of claim 1 , wherein the topics are at least one of an artist name, a song name, a product name, a user-assigned tag, and a Freebase entity.

3. The system of claim 1 , wherein the distance metric is a mutual information score.

4. The system of claim 1 , wherein the span is two adjacent presentations.

5. The system of claim 1 , wherein the single rank-ordered list of all second topics generated for each of the plurality of pieces of media content is generated by:
(a) determining topics associated with the piece of media content;
(b) for each of the topics, obtaining a rank-ordered list of second topics generated for the topic as a first topic; and
(c) combining all of the rank-ordered lists obtained at (b) together using the distance metric as ranking score to form the single rank-ordered list.

6. The system of claim 1 , wherein the single rank-ordered list of all second topics generated for each of the plurality of pieces of media content is generated by:
(a) determining all possible subsequent topics for the piece of media content;
(b) for each of the possible subsequent topics (“c”):
(i) for each (“k1”) of the topics associated with the piece of media content, determine the probability that a presentation with topic c will follow a presentation with topic k1, and then multiply these probabilities together to form a value X,
(ii) for each (“k2”) of the topics associated with the piece of media content, determine the probability that a presentation with topic c will not follow a presentation with topic k2, and then multiply these probabilities together to form a value Y,
(iii) calculate Z=X/(X÷Y), and
(iv) assign the value of Z to the particular value c; and

(c) forming a rank-ordered list of the different values of c, wherein the values of c are ranked using the corresponding values of Z as a ranking score.

7. The system of claim 1 , wherein the one or more other pieces of media content identified as recommended media content based on the single rank-ordered list for each of the plurality of pieces of media content is identified by:
(a) for each other piece of media content:
(i) identifying all topics associated with the other piece of media content,
(ii) averaging the ranking scores for the identified topics for the other piece of media content to form an average, and
(iii) assigning the average as a recommendation score for the other piece of media content; and

(b) selecting N other pieces of media content with top recommendation scores as the recommended media content for the piece of media content.

8. A method for recommending media content, comprising:
identifying, using a hardware processor, topics associated with a plurality of pieces of media content presented in a session;
calculating, using the hardware processor, a distance metric for pairs of topics, wherein each of the pairs of topics includes a first topic associated with a first piece of media content and a second topic associated with a second piece of media content, and wherein the second piece of media content was presented within a given span of the presentation of the first piece of media content;
for each of the plurality of pieces of media content, generating, using the hardware processor, a single rank-ordered list of all second topics; and
for each of the plurality of pieces of media content, identifying, using the hardware processor, one or more other pieces of media content as recommended media content based on the single rank-ordered list.

9. The method of claim 8 , wherein the topics are at least one of an artist name, a song name, a product name, a user-assigned tag, and a Freebase entity.

10. The method of claim 8 , wherein the distance metric is a mutual information score.

11. The method of claim 8 , wherein the span is two adjacent presentations.

12. The method of claim 8 , further comprising generating the single rank-ordered list of all second topics for each of the plurality of pieces of media content by:
(a) determining topics associated with the piece of media content;
(b) for each of the topics, obtaining a rank-ordered list of second topics generated for the topic as a first topic; and
(c) combining all of the rank-ordered lists obtained at (b) together using the distance metric as ranking score to form the single rank-ordered list.

13. The method of claim 8 , further comprising generating the single-rank ordered list of all second topics for each of the plurality of pieces of media content by:
(a) determining all possible subsequent topics for the piece of media content;
(b) for each of the possible subsequent topics (“c”):
(i) for each (“k1”) of the topics associated with the piece of media content, determine the probability that a presentation with topic c will follow a presentation with topic k1, and then multiply these probabilities together to form a value X,
(ii) for each (“k2”) of the topics associated with the piece of media content, determine the probability that a presentation with topic c will not follow a presentation with topic k2, and then multiply these probabilities together to form a value Y,
(iii) calculate Z=X/(X÷Y), and
(iv) assign the value of Z to the particular value c; and

(c) forming a rank-ordered list of the different values of c, wherein the values of c are ranked using the corresponding values of Z as a ranking score.

14. The method of claim 8 , further comprising identifying the one or more other pieces of media content as recommended media content based on the single rank-ordered list for each of the plurality of pieces of media content by:
(a) for each other piece of media content:
(i) identifying all topics associated with the other piece of media content,
(ii) averaging the ranking scores for the identified topics for the other piece of media content to form an average, and
(iii) assigning the average as a recommendation score for the other piece of media content; and

(b) selecting N other pieces of media content with top recommendation scores as the recommended media content for the piece of media content.

15. A non-transitory computer-readable medium containing computer-executable instructions that, when executed by a processor, cause the processor to perform a method for recommending media content, the method comprising:
identifying topics associated with a plurality of pieces of media content presented in a session;
calculating a distance metric for pairs of topics, wherein each of the pairs of topics includes a first topic associated with a first piece of media content and a second topic associated with a second piece of media content, and wherein the second piece of media content was presented within a given span of the presentation of the first piece of media co

Assignees

• Google Inc

Inventors

• Kumthekar Aditee
• He Yu
• Murphy Kevin Patrick

Classifications

• G06F17/30023

  Physics · mapped topic

• H04N21/4826 Primary

  using recommendation lists, e.g. of programmes or channels sorted out according to their score · CPC title

• G06F17/30864

  Physics · mapped topic

• H04N21/442

  Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk {(arrangements for monitoring broadcast services or broadcast-related services H04H60/29; arrangements for identifying or recognising characteristics with a direct linkage to broadcast information H04H60/35; monitoring of user activities for profile generation for accessing a video database G06F16/739; monitoring in wireless networks H04W24/00)} · CPC title

• H04N21/4828

  for searching programme descriptors (retrieval of video data G06F16/739) · CPC title