Context based systems and methods for presenting media file annotation recommendations

• US 9,678,993 B2
• Filed: 01/28/2014
• Issued: 06/13/2017
• Est. Priority Date: 03/14/2013
• Status: Active Grant

• Current Assignee: Shutterstock Incorporated
• Original Assignee: Shutterstock Incorporated
• Inventors: Hohwald, Heath Walter
• Primary Examiner(s): Bullock, Joshua
• Application Number:
  US14/165,694
• Publication Number:
  US 20140280113A1
• Time in Patent Office: 1,232 Days
• Field of Search: 707728
• CPC Class Codes: G06F 16/5866 using information manually generated, e.g. tags, keywords, comments, manually generated location and time information

Abstract

A context based tag recommendation system for annotating stock photography media files, the system configured to: maintain a first database comprising a set of media files and a set of annotations associated with the media file set, maintain a second database comprising a second set of records, each record comprising an annotation, at least one co-occurring annotation, and a count, receive at least one media file tag input (input), query the second database using the input to identify a ranked set of co-occurring annotations and a count associated with each annotation in the ranked set of co-occurring annotations, query the first database to identify a selected set of similar media files, by iteratively retrieving image files associated a selective group of annotations from the ranked set of co-occurring annotations, present a list of recommended annotations based on the ranked set of co-occurring annotations and the selected set of similar media files.

Claims

1. A context based tag recommendation system for precisely annotating stock photography media files, the system comprising:

a memory comprising instructions; and

a data processor executing the instructions to perform the following operations;

receive, for an input media file, an input annotation;

identify, in a second database, (1) a set of co-occurring annotations that occur for media files associated with the input annotation, and (2) a count associated with each co-occurring annotation from the set of annotations indicating for how many media files in a first database the co-occurring annotation occurs with the input annotation, wherein the second database comprises records, each record in the comprising the count indicating the number of media files from the first database associated with the annotation and the respective co-occurring annotation;

identify, in a first database, media files similar to the input media file by retrieving media files associated with annotations, from the set of co-occurring annotations, having an associated count meeting a predefined criterion indicating a minimum number of media files in the first database for which the co-occurring annotation occurs with the input annotation; and

selecting annotations associated with the similar media files that have the associated count meeting the predefined criterion to present as a list of recommended annotations for the input media file.

2. The system of claim 1, wherein the predefined criterion further comprises media files that have appeared together in a same query.

3. The system of claim 1, wherein the predefined criterion further comprises media files that have appeared in a same download transaction.

4. The system of claim 1, wherein the input annotation is provided by a media file contributor.

5. The system of claim 4, wherein each annotation in the set of co-occurring annotations is separate and distinct from the input annotation provided by the media file contributor.

6. The system of claim 1, wherein the input annotation is provided by a user.

7. The system of claim 1, wherein the system is further configured to maintain in a transaction history title information and behavioral data associated with the similar media file files.

8. The system of claim 7, wherein the transaction history comprises, for each received input annotation, the set of co-occurring annotations and identified similar media files.

9. The system of claim 7, wherein the transaction history comprises, for each received input annotation, media files downloaded from the first database associated with the input annotation.

10. A method comprising:

receiving, for an input media file, an input annotation;

identifying, in a second database, (1) a set of co-occurring annotations that occur for media files associated with the input annotation, and (2) a count associated with each co-occurring annotation from the set of annotations indicating for how many media files in a first database the co-occurring annotation occurs with the input annotation, wherein the second database comprises records, each record in the comprising the count indicating the number of media files from the first database associated with the annotation and the respective co-occurring annotation;

identifying, in a first database, media files similar to the input media file by retrieving media files associated with annotations, from the set of co-occurring annotations, having an associated count meeting a predefined criterion indicating a minimum number of media files in the first database for which the co-occurring annotation occurs with the input annotation; and

selecting annotations associated with the similar media files that have the associated count meeting the predefined criterion to present as a list of recommended annotations for the input media file.

11. The method of claim 10, wherein the predefined criterion further comprises media files that have appeared together in a same query.

12. The method of claim 10, wherein the predefined criterion further comprises media files that have appeared in a same download transaction.

13. The method of claim 10, wherein the input annotation is provided by a media file contributor.

14. The method of claim 13, wherein each annotation in the set of co-occurring annotations is separate and distinct from the input annotation provided by the media file contributor.

15. The method of claim 10, wherein the input annotation is provided by a user.

16. The method of claim 10, wherein the method further comprises maintaining in a transaction history title information and behavioral data associated with the similar media file files.

17. The method of claim 16, wherein the transaction history comprises, for each received input annotation, the set of co-occurring annotations and identified similar media files.

18. The method of claim 16, wherein the transaction history comprises, for each received input annotation, media files downloaded from the first database associated with the input annotation.