Method and system for populating a concept database with respect to user identifiers

• US 10,585,934 B2
• Filed: 07/12/2017
• Issued: 03/10/2020
• Est. Priority Date: 10/26/2005
• Status: Active Grant

• Current Assignee: Cortica Ltd.
• Original Assignee: Cortica Ltd.
• Inventors: Raichelgauz, Igal, Odinaev, Karina, Zeevi, Yehoshua Y
• Primary Examiner(s): Rostami, Mohammad S
• Application Number:
  US15/647,888
• Publication Number:
  US 20170315994A1
• Time in Patent Office: 972 Days
• CPC Class Codes: G06F 16/14 Details of searching files based on file metadata
• CPC Class Codes: G06F 16/152 using file content signatures, e.g. hash values
• CPC Class Codes: G06F 16/41 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface
• CPC Class Codes: Y10S 707/99948 Application of database or data structure, e.g. distributed, multimedia, or image

Abstract

A method and system for populating a concept database based on a plurality of multimedia data elements (MMDEs) and at least one user identifier, the concept database including a plurality of concept structures. The method includes clustering a plurality of signatures of the plurality of MMDEs into a cluster of signatures; reducing the number of signatures in the cluster to a create a signature reduced cluster (SRC); generating, based on the SRC and the plurality of concept structures, an index, wherein the index maps the plurality of MMDEs to a matching concept structure of the plurality of concept structures; associating the index with a user based on the at least one user identifier; and storing the index and the plurality of MMDEs in the concept database.

Claims

1. A method for populating a concept database based on a plurality of multimedia data elements (MMDEs) and at least one user identifier, the concept database including a plurality of concept structures, comprising:

clustering a plurality of signatures of the plurality of MMDEs into a cluster of signatures;

reducing the number of signatures in the cluster to create a signature reduced cluster (SRC);

generating, based on the SRC and the plurality of concept structures, an index, wherein the index maps the plurality of MMDEs to a matching concept structure of the plurality of concept structures;

associating the index with a user based on the at least one user identifier;

storing the index and the plurality of MMDEs in the concept database;

finding a first SRC and a second SRC that are associated to each other;

wherein the finding comprises determining that the first SRC is associated with the second SRC based on (a) a partial match between metadata associated with first SRC and metadata associated with the second SRC, and (b) a partial match between one or more signatures of the first SRC and one or more signatures of the second SRC; and

associating, with the second SRC, metadata related to the first SRC.

2. The method of claim 1, wherein each user identifier is a visual identifier showing one of at least one user, wherein associating the user with the index further comprises:

comparing at least one signature of each of the at least one user identifier to a signature of the matching concept structure to determine a matching user identifier, wherein the associated user is the user shown in the matching user identifier.

3. The method of claim 2, further comprising:

enabling access to the plurality of MMDEs in the concept database by at least one user device of the associated user.

4. The method of claim 2, further comprising:

retrieving the at least one user identifier from at least one web platform;

sending the plurality of MMDEs to the associated user via the at least one web platform.

5. The method of claim 2, wherein generating the index for each created SRC further comprises:

generating metadata for each signature of the created SRC;

comparing the generated metadata to metadata associated with the plurality of concept structures to determine the matching concept structure, wherein each concept structure includes a plurality of SRCs and associated metadata.

6. The method of claim 5, further comprising:

removing duplicated metadata of the generated metadata.

7. The method of claim 5, further comprising:

for each of at least one portion of the generated metadata, removing all instances of the portion from the generated metadata when a number of repetitions of the portion in the generated metadata is below a threshold.

8. The method of claim 1, wherein clustering the plurality of signatures further comprises:

generating a plurality of items from a received MMDE of the plurality of MMDEs;

determining the items that are of interest for signature generation; and

sending, to a signature generator, the determined items of interest, wherein the signature generator is configured to generate the plurality of signatures based on the determined items of interest.

9. The method according to claim 1 comprising generating the plurality of signatures of the plurality of MMDEs;

wherein each signature of the plurality of signatures is a response of one or more neural networks to at least a portion of one of the MMDEs.

10. The method according to claim 2, comprising generating the plurality of signatures of the plurality of MMDEs by independent computational cores.

11. A non-transitory computer readable medium having stored thereon instructions for causing one or more processing units to execute a process, the process comprising:

clustering a plurality of signatures of a plurality of MMDEs into a cluster of signatures;

reducing the number of signatures in the cluster to a create a signature reduced cluster (SRC);

generating, based on the SRC and a plurality of concept structures stored in a concept database, an index, wherein the index maps the plurality of MMDEs to a matching concept structure of the plurality of concept structures;

associating the index with a user based on the at least one user identifier;

storing the index and the plurality of MMDEs in the concept databases;

finding a first SRC and a second SRC that are associated to each other;

wherein the finding comprises determining that the first SRC is associated with the second SRC based on (a) a partial match between metadata associated with first SRC and metadata associated with the second SRC, and (b) a partial match between one or more signatures of the first SRC and one or more signatures of the second SRC; and

associating, with the second SRC, metadata related to the first SRC.

12. A system for populating a concept database based on a plurality of multimedia data elements (MMDEs) and at least one user identifier, the concept database including a plurality of concept structures, comprising:

a processing circuitry; and

a memory connected to the processing circuitry, the memory containing instructions that when executed by the processing circuitry, configure the system to;

cluster a plurality of signatures of the plurality of MMDEs into a cluster of signatures;

reduce the number of signatures in the cluster to create a signature reduced cluster (SRC);

generate, based on the SRC and the plurality of concept structures, an index, wherein the index maps the plurality of MMDEs to a matching concept structure of the plurality of concept structures;

associate the index with a user based on the at least one user identifier;

store the index and the plurality of MMDEs in the concept database;

find a first SRC and a second SRC that are associated to each other by determining that the first SRC is associated with the second SRC based on (a) a partial match between metadata associated with first SRC and metadata associated with the second SRC, and (b) a partial match between one or more signatures of the first SRC and one or more signatures of the second SRC; and

associate, with the second SRC, metadata related to the first SRC.

13. The system of claim 12, wherein each user identifier is a visual identifier showing one of at least one user, wherein the system is further configured to:

compare at least one signature of each of the at least one user identifier to a signature of the matching concept structure to determine a matching user identifier, wherein the associated user is the user shown in the matching user identifier.

14. The system of claim 13, wherein the system is further configured to:

enable access to the plurality of MMDEs in the concept database by at least one user device of the associated user.

15. The system of claim 13, wherein the system is further configured to:

retrieve the at least one user identifier from at least one web platform;

send the plurality of MMDEs to the associated user via the at least one web platform.

16. The system of claim 13, wherein the system is further configured to:

generate metadata for each signature of the created SRC;

compare the generated metadata to metadata associated with the plurality of concept structures to determine the matching concept structure, wherein each concept structure includes a plurality of SRCs and associated metadata.

17. The system of claim 16, wherein the system is further configured to:

remove duplicated metadata of the generated metadata.

18. The system of claim 16, wherein the system is further configured to:

for each of at least one portion of the generated metadata, remove all instances of the portion from the generated metadata when a number of repetitions of the portion in the generated metadata is below a threshold.

19. The system of claim 12, wherein the system is further configured to:

generate a plurality of items from a received MMDE of the plurality of MMDEs;

determine the items that are of interest for signature generation; and

send, to a signature generator, the determined items of interest, wherein the signature generator is configured to generate the plurality of signatures based on the determined items of interest.

20. A method for populating a concept database based on a plurality of multimedia data elements (MMDEs) and at least one user identifier, the concept database including a plurality of concept structures, comprising:

generating a plurality of signatures of the plurality of MMDEs;

wherein each signature of the plurality of signatures is a response of one or more neural networks to at least a portion of one of the MMDEs clustering a plurality of signatures of the plurality of MMDEs into a cluster of signatures;

reducing the number of signatures in the cluster to create a signature reduced cluster (SRC);

generating, based on the SRC and the plurality of concept structures, an index, wherein the index maps the plurality of MMDEs to a matching concept structure of the plurality of concept structures; and

associating the index with a user based on the at least one user identifier;

storing the index and the plurality of MMDEs in the concept database;

wherein each user identifier is a visual identifier showing one of at least one user, wherein associating the user with the index further comprises;

comparing at least one signature of each of the at least one user identifier to a signature of the matching concept structure to determine a matching user identifier, wherein the associated user is the user shown in the matching user identifier;

generating the plurality of signatures of the plurality of MMDEs by independent computational cores;

wherein each of said independent computational cores comprising properties having at least some statistical independency from other of said computational cores, said properties being set independently of each other of said computational cores.

21. The method according to claim 20 comprising finding a first SRC and a second SRC that are associated to each other;

and associating, with the second SRC, metadata related to the first SRC.

22. The method according to claim 21 comprising determining that the first SRC is associated with the second SRC based on (a) a partial match between metadata associated with first SRC and metadata associated with the second SRC, and (b) a partial match between one or more signatures of the first SRC and one or more signatures of the second SRC.

23. The method according to claim 21 comprising associating the second SRC with a concept structure associated with the first SRC.