Methods and systems for updating database tags for media content

• US 10,198,498 B2
• Filed: 05/13/2015
• Issued: 02/05/2019
• Est. Priority Date: 05/13/2015
• Status: Active Grant

• Current Assignee: ROVI Product Corporation (Xperi Inc.)
• Original Assignee: Rovi Guides, Inc. (Adeia Inc.)
• Inventors: Venkataraman, Sashikumar, Agarwal, Vineet, Mohaideen P, Ahmed Nizam
• Primary Examiner(s): Saeed, Usmaan
• Assistant Examiner(s): Nafziger, David M
• Application Number:
  US14/710,965
• Publication Number:
  US 20160335339A1
• Time in Patent Office: 1,364 Days
• CPC Class Codes: G06F 16/23 Updating
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/48 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/738 Presentation of query results
• CPC Class Codes: G06F 16/78 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• 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 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results

Abstract

Methods and systems for a media guidance application that can automatically, or with little human interaction, identify characteristics of media content (e.g., sub-genres, minor actors, etc.) and generate appropriate tags for that media content. The media guidance application can thus facilitate user searches based on specific and narrowly tailored search terms (e.g., sub-genres, minor actors, etc.).

Claims

1. A method for providing detailed tagging for media content searches by updating database tags for media content based on a comparison of known characteristics of media assets corresponding to a selected tag to characteristics of media assets not currently corresponding to the selected tag, the method comprising:

selecting a media asset record from a first database, wherein the first database lists a plurality of available media asset records for a media asset, wherein each of the plurality of available media asset records is a descriptive metadata tag;

processing the selected media asset record to determine a plurality of keywords corresponding to the selected media asset record based on a respective descriptive metadata tag for the selected media asset record;

storing a plurality of keyword vectors in a second database, wherein each of the plurality of keyword vectors indicates a meaning of a respective keyword based on a frequency at which other words surround the respective keyword, and wherein a dot product between two keyword vectors of the plurality of keyword vectors indicates a level of similarity between respective keywords of the two keyword vectors;

determining a respective dot product for each keyword vector of the plurality of keyword vectors and each of the remaining keyword vectors of the plurality of keyword vectors;

comparing the respective dot product for each keyword vector of the plurality of keyword vectors to a threshold value;

in response to determining which of the respective dot products exceeds the threshold value, determining a subject matter vector based on the similar keyword vectors with the respective dot products that exceed the threshold value, wherein the subject matter vector indicates keywords corresponding to the subject matter;

comparing the subject matter vector to an attribute vector to determine whether the subject matter vector and the attribute vector correspond, wherein the attribute vector includes a plurality of vectors associated with a selected attribute, and wherein the selected attribute is a characteristic of the media asset used to distinguish the media asset from other media assets;

in response to determining that the subject matter vector and the attribute vector correspond, tagging the selected media asset record with the selected attribute in the first database;

receiving a user request for media assets corresponding to the selected attribute; and

generating for display a listing for the selected media asset record based on the tagging.

2. The method of claim 1, wherein each of the keywords in the plurality of keywords is identified from subtitle data or descriptive textual data associated with the media asset.

3. The method of claim 1, further comprising:

associating a keyword vector weight with each of the two keyword vectors based on a keyword weight corresponding to the respective keywords; and

determining the subject matter vector based on the keyword vector weight associated with each of the two keyword vectors.

4. The method of claim 1, wherein a weight associated with each of the plurality of vectors associated with the selected attribute is used to determine whether the subject matter vector and the attribute vector correspond.

5. The method of claim 1, wherein comparing the respective dot product for each keyword vector of the plurality of keyword vectors to a threshold value comprises:

determining a first dot product of a first keyword vector of the plurality of keyword vectors and a second keyword vector of the plurality of keyword vectors;

retrieving a first threshold value from the second database; and

comparing the dot product to the first threshold value.

6. The method of claim 1, wherein the attribute vector includes a plurality of values associated with a selected attribute, and wherein comparing the subject matter vector to an attribute vector to determine whether the subject matter vector and the attribute vector correspond further comprises:

determining a second dot product of the subject matter vector and the attribute vector;

retrieving a second threshold value from the second database; and

comparing the second dot product to the second threshold value.

7. The method of claim 1, wherein the selected attribute is a genre.

8. The method of claim 1, further comprising weighting the subject matter vector based on a number of keyword vectors of the plurality of keyword vectors that correspond to the subject matter vector.

9. The method of claim 8, further comprising determining whether to compare the subject matter vector to the attribute vector based on a weight corresponding to the subject matter vector.

10. A system for providing detailed tagging for media content searches by updating database tags for media content based on a comparison of known characteristics of media assets corresponding to a selected tag to characteristics of media assets not currently corresponding to the selected tag, the system comprising:

storage circuitry configured to store a first database, wherein the first database lists a plurality of available media asset records for a media asset, wherein each of the plurality of available media asset records is a descriptive metadata tag; and

control circuitry configured to;

select a media asset record from the first database;

process the selected media asset record to determine a plurality of keywords corresponding to the selected media asset record based on a respective descriptive metadata tag for the selected media asset record;

store a plurality of keyword vectors in a second database, wherein each of the plurality of keyword vectors indicates a meaning of a respective keyword based on a frequency at which other words surround the respective keyword, and wherein a dot product between two keyword vectors of the plurality of keyword vectors indicates a level of similarity between respective keywords of the two keyword vectors;

determining a respective dot product for each keyword vector of the plurality of keyword vectors and each of the remaining keyword vectors of the plurality of keyword vectors;

compare the respective dot product for each keyword vector of the plurality of keyword vectors to a threshold value;

in response to determining which of the respective dot products exceeds the threshold value, determine a subject matter vector based on the keyword vectors with the respective dot products that exceed the threshold value, wherein the subject matter vector indicates keywords corresponding to the subject matter;

compare the subject matter vector to an attribute vector to determine whether the subject matter vector and the attribute vector correspond, wherein the attribute vector includes a plurality of vectors associated with a selected attribute, and wherein the selected attribute is a characteristic of the media asset used to distinguish the media asset from other media assets;

in response to determining that the subject matter vector and the attribute vector correspond, tag the selected media asset record with the selected attribute in the first database;

receive a user request for media assets corresponding to the selected attribute; and

generate for display a listing for the selected media asset record based on the tagging.

11. The system of claim 10, wherein each of the keywords in the plurality of keywords is identified from subtitle data or descriptive textual data associated with the media asset.

12. The system of claim 10, wherein the control circuitry is further configured to:

associate a keyword vector weight with each of the two keyword vectors based on a keyword weight corresponding to the respective keywords; and

determine the subject matter vector based on the keyword vector weight associated with each of the two keyword vectors.

13. The system of claim 10, wherein a weight associated with each of the plurality of vectors associated with the selected attribute is used to determine whether the subject matter vector and the attribute vector correspond.

14. The system of claim 10, wherein the control circuitry configured to compare the respective dot product for each keyword vector of the plurality of keyword vectors to a threshold value is further configured to:

determine a first dot product of a first keyword vector of the plurality of keyword vectors and a second keyword vector of the plurality of keyword vectors;

retrieve a first threshold value from the second database; and

compare the first dot product to the first threshold value.

15. The system of claim 10, wherein the attribute vector includes a plurality of values associated with a selected attribute, and wherein the control circuitry configured to compare the subject matter vector to an attribute vector to determine whether the subject matter vector and the attribute vector correspond is further configured to:

determine a second dot product of the subject matter vector and the attribute vector;

retrieve a second threshold value from the second database; and

compare the second dot product to the second threshold value.

16. The system of claim 10, wherein the selected attribute is a genre.

17. The system of claim 10, wherein the control circuitry is further configured to weight the subject matter vector based on a number of keyword vectors of the plurality of keyword vectors that correspond to the subject matter vector.

18. The system of claim 17, wherein the control circuitry is further configured to determine whether to compare the subject matter vector to the attribute vector based on a weight corresponding to the subject matter vector.