System and method for identifying media content items and related media content items

• US 8,166,029 B2
• Filed: 09/07/2006
• Issued: 04/24/2012
• Est. Priority Date: 09/07/2006
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Park, Seung-Taek, Pennock, David Myer
• Primary Examiner(s): Hu, Jensen
• Application Number:
  US11/518,378
• Publication Number:
  US 20080126303A1
• Time in Patent Office: 2,056 Days
• Field of Search: 707/102, 707/732
• US Class Current: 707/732
• CPC Class Codes: G06F 16/735 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 16/78 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually

Abstract

A method comprising receiving a search query to generate a search result of one or more media items and providing a personalized search rank of the one or more media items on the basis of a user profile and an item relevance for a given media item with regard to the query metadata associated with the search result is identified and used to identify at least one related media item. The at least one related media item is ranked on the basis of the user profile and the metadata.

Claims

1. A computerized method comprising:

receiving a search query to generate a search result of one or more media items;

providing a personalized search rank of the one or more media items on the basis of a user profile and an item relevance for a given media item with regard to the query, the user profile including genre preferences, actor preferences, search histories, search result interactions, and items viewed, ordered or utilized by a user of the user profile;

identifying, via a processing device, metadata associated with the search result;

using the metadata associated with the search result to identify, via the processing device, a set of related media items, the related media items identified based on a movie-casting graph and a movie-user graph, the movie-casting graph including associations between one or more movies and one or more actors, and the movie-user graph including associations between one or more movies rated by the user, wherein the movie-casting and the movie-user graphs are generated using metadata associated with the one or more media items;

generating one or more unigram scores by counting a number of unigrams in the search query that appear in or match one or more metadata fields associated with the one or more media items, wherein a given metadata field comprises metadata identifying a given media item;

generating one or more bigram scores by counting a number of bigrams in the search query that appear in or match one or more metadata fields associated with the one or more media items;

generating a proximity score for the one or more media items by combining the one or more bigram scores and the one or more unigram scores, the bigram score having a greater weight than the unigram score in generating the proximity score;

generating an authority value for the one or more media items, wherein the authority value is based on a personalized search rank of the one or more media items, the personalized search rank of the one or more media items including, for each of the one or more media items, an explicit rating, a predicted rating based on the user profile, and a global popularity, wherein one or more weights are assigned to the explicit rating, the predicted rating and the global popularity based on the user profile in generating the authority value;

combining the proximity score and the authority value for the one or more media items;

ranking, via the processing device, the one or more media items based on the combined proximity score and authority value;

providing the ranked one or more media items in the search result along with the identified set of related media items.

2. The method of claim 1 further comprising tracking activity on the media items of the search result, and storing the tracked data in a preference profile.

3. The method of claim 1 wherein the personalized search rank is determined according to a preference profile.

4. The method of claim 1 wherein the item relevance comprises a web relevance.

5. The method of claim 1 wherein the item relevance comprises a DB relevance.

6. The method of claim 1 wherein the at least one related media item comprises one or more movies connected by common cast and crewmembers.

7. The method of claim 1 wherein the at least one related media item comprises one or more movies connected by item similarities in user preferences.

8. Non-transitory computer readable media comprising:

a search engine operative to receive a search query and generating a search result of one or more media items, the search result organized in accordance with a personalized search rank on the basis of a user profile and an item relevance for a given media item with regard to the query, the user profile including genre preferences, actor preferences, search histories, search result interactions, and items viewed, ordered or utilized by a user of the user profile;

a user preference agent coupled to the search engine and operative to identify metadata associated with the search result and operative to use the metadata associated with the search result to identify a set of related media items, the related media items identified based on a movie-casting graph and a movie-user graph, the movie-casting graph including associations between one or more movies and one or more actors, and the movie-user graph including associations between one or more movies rated by the user, wherein the movie-casting graph and the movie-user graphs generated by a graph generator using metadata associated with the one or more media items;

a score generator operative to generate one or more unigram scores by counting a number of unigrams in the search query that appear in or match the one or more metadata fields associated with the one or more media items, wherein a given metadata field comprises metadata identifying a given media item;

a score generator operative to generate one or more bigram scores by counting a number of bigrams in the search query that appear in or match in the one or more metadata fields associated with the one or more media item;

a score generator operative to generate a proximity score for the one or more media items by combining the one or more bigram scores and the one or more unigram scores, the bigram score having a greater weight than the unigram score in generating the proximity score;

a value generator operative to generate an authority value, wherein the authority value is based on a personalized search rank of the one or more media items, the personalized search rank of the one or more media items including, for each of the one or more media items, an explicit rating, a predicted rating based on the user profile, and a global popularity, wherein one or more weights are assigned to the explicit rating, the predicted rating and the global popularity based on the user profile in generating the authority value;

a combiner operative to combine the proximity score and the authority value; and

a ranking component operative to rank, via the processing device, the one or more media items based on the combined proximity score and authority value; and

an interface component operative to provide the ranked one or more media items in the search result along with the identified set of related media items.

9. The non-transitory computer readable media of claim 8 further comprising interface components for tracking activity on the media items of the search result, and storing the tracked data in a preference profile.

10. The non-transitory computer readable media of claim 8 wherein the personalized search rank is determined in accordance with a preference profile.

11. The non-transitory computer readable media of claim 8 wherein the item relevance comprises a web relevance.

12. The non-transitory computer readable media of claim 8 wherein the item relevance comprises a DB relevance.

13. The non-transitory computer readable media of claim 8 wherein the at least one related media item comprises one or more movies connected by at least one common cast and crewmember.

14. The non-transitory computer readable media of claim 8 wherein the at least one related media item comprises one or more movies connected by item similarities based on user preferences.

15. A system comprising:

a memory device having executable instructions stored therein; and

a processing device, in response to the executable instructions, operative to;

receive a search query to generate a search result of one or more media items;

provide a personalized search rank of the one or more media items on the basis of a user profile and an item relevance for a given media item with regard to the query, the user profile including genre preferences, actor preferences, search histories, search result interactions, and items viewed, ordered or utilized by a user of the user profile;

identify metadata associated with the search result;

use the metadata associated with the search result to identify a set of related media items, the related media items identified based on a movie-casting graph and a movie-user graph, the movie-casting graph including associations between one or more movies and one or more actors, and the movie-user graph including associations between one or more movies rated by the user, wherein the movie-casting graph and the movie-user graphs are generated using metadata associated with the one or more media items;

generate one or more unigram scores by counting a number of unigrams in the search query that appear in or match one or more metadata fields associated with the one or more media items, wherein a given metadata field comprises metadata identifying a given media item;

generate one or more bigrams score by counting a number of bigrams in the search query that appear in or match the one or more metadata fields associated with the one or more media items;

generate a proximity score for the one or more media items by combining the one or more bigram scores and the one or more unigram scores, the bigram score having a greater weight than the unigram score in generating the proximity score;

generate an authority value, wherein the authority value is based on a personalized search rank of the one or more media items, the personalized search rank of the one or more media items including, for each of the one or more media items, an explicit rating, a predicted rating based on the user profile, and a global popularity, wherein one or more weights are assigned to the explicit rating, the predicted rating and the global popularity based on the user profile in generating the authority value;

combine the proximity score and the authority value;

rank, via the processing device, the one or more media items based on the combined proximity score and authority value; and

provide the ranked one or more media items in the search result along with the identified set of related media items.