Granted Patent B2

US 12,273,594 · App. 17/248,539 · Granted Apr 8, 2025

Inventors: Alessio Tamburro (Wilmington, DE); Meenakshi Sundaram Bagavathi Krishnan (Philadelphia, PA); Venkata Gunnu (Malvern, PA)

Assignee: Comcast Cable Communications, LLC

H04N21/4668H04N21/44204H04N21/4662

Abstract

Claims (96)

1. A method comprising:

determining a request associated with a user for a content recommendation;

determining, based on a viewing history associated with the user, at least one content item comprising video content;

determining, based on comparing encoded data associated with text associated with the at least one content item with encoded data associated with the text associated with each candidate content item of a plurality of candidate content items, a subset of candidate content items of the plurality of candidate content items, wherein the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item each comprise one or more vectors configured to facilitate natural language processing;

determining, based on comparing encoded data associated with at least one image associated with the at least one content item with encoded data associated with at least one image associated with each of the subset of candidate content items, at least one candidate content item of the subset of candidate content items, wherein the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each of the subset of candidate content items each comprise one or more vectors configured to facilitate image classification or detection; and

sending, to a device associated with the user, an indication of the at least one candidate content item.

2. The method of claim 1 , further comprising:

determining the encoded data associated with the text associated with each of the plurality of candidate content items using a first machine learning algorithm; and

determining the encoded data associated with the at least one image associated with each of the subset of candidate content items using a second machine learning algorithm.

3. The method of claim 1 , wherein determining the subset of candidate content items comprises:

determining a similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items.

4. The method of claim 3 , wherein determining the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items comprises:

determining a cosine similarity distance between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items.

5. The method of claim 3 , wherein determining the subset of candidate content items further comprises:

comparing the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items with a threshold; and

determining that the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with the at least one candidate content item satisfies the threshold.

6. The method of claim 1 , wherein determining the at least one candidate content item comprises:

determining a similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items.

7. The method of claim 6 , wherein determining the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items comprises:

determining a cosine similarity distance between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items.

8. The method of claim 6 , wherein determining the at least one candidate content item further comprises:

comparing the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items with a threshold; and

determining that the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with the at least one candidate content item satisfies the threshold.

9. A device comprising:

one or more processors; and

memory storing instructions that, when executed by the one or more processors, cause the device to:

determine a request associated with a user for a content recommendation;

determine, based on a viewing history associated with the user, at least one content item comprising video content;

determine, based on comparing encoded data associated with text associated with the at least one content item with encoded data associated with text associated with each candidate content item of a plurality of candidate content items, a subset of candidate content items of the plurality of candidate content items, wherein the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item each comprise one or more vectors configured to facilitate natural language processing;

determine, based on comparing encoded data associated with at least one image associated with the at least one content item with encoded data associated with at least one image associated with each of the subset of candidate content items, at least one candidate content item of the subset of candidate content items, wherein the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each of the subset of candidate content items each comprise one or more vectors configured to facilitate image classification or detection; and

send, to an output device associated with the user, an indication of the at least one candidate content item.

10. The device of claim 9 , wherein the instructions, when executed by the one or more processors, further cause the device to:

determine the encoded data associated with the text associated with each of the plurality of candidate content items using a first machine learning algorithm; and

determine the encoded data associated with the at least one image associated with each of the subset of candidate content items using a second machine learning algorithm.

11. The device of claim 9 , wherein the instructions that, when executed by the one or more processors, cause the device to determine the subset of candidate content items comprise instructions that, when executed by the one or more processors, cause the device to:

determine a similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items.

12. The device of claim 11 , wherein the instructions that, when executed by the one or more processors, cause the device to determine the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items comprise instructions that, when executed by the one or more processors, cause the device to:

determine a cosine similarity distance between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items.

13. The device of claim 11 , wherein the instructions that, when executed by the one or more processors, cause the device to determine the subset of candidate content items further comprise instructions that, when executed by the one or more processors, cause the device to:

compare the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items with a threshold; and

determine that the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with the at least one candidate content item satisfies the threshold.

14. The device of claim 9 , wherein the instructions that, when executed by the one or more processors, cause the device to determine the at least one candidate content item comprise instructions that, when executed by the one or more processors, cause the device to:

determine a similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items.

15. The device of claim 14 , wherein the instructions that, when executed by the one or more processors, cause the device to determine the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items comprise instructions that, when executed by the one or more processors, cause the device to:

determine a cosine similarity distance between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items.

16. The device of claim 14 , wherein the instructions that, when executed by the one or more processors, cause the device to determine the at least one candidate content item further comprise instructions that, when executed by the one or more processors, cause the device to:

compare the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items with a threshold; and

determine that the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with the at least one candidate content item satisfies the threshold.

17. A computer-readable medium storing instructions that, when executed, cause:

determining a request associated with a user for a content recommendation;

determining, based on a viewing history associated with the user, at least one content item comprising video content;

determining, based on comparing encoded data associated with text associated with the at least one content item with encoded data associated with text associated with each candidate content item of a plurality of candidate content items, a subset of candidate content items of the plurality of candidate content items, wherein the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item each comprise one or more vectors configured to facilitate natural language processing;

determining, based on comparing encoded data associated with at least one image associated with the at least one content item with encoded data associated with at least one image associated with each of the subset of candidate content items, at least one candidate content item of the subset of candidate content items, wherein the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each of the subset of candidate content items each comprise one or more vectors configured to facilitate image classification or detection; and

sending, to a device associated with the user, an indication of the at least one candidate content item.

18. The computer-readable medium of claim 17 , wherein the instructions, when executed, further cause:

determining the encoded data associated with the text associated with each of the plurality of candidate content items using a first machine learning algorithm; and

determining the encoded data associated with the at least one image associated with each of the subset of candidate content items using a second machine learning algorithm.

19. The computer-readable medium of claim 17 , wherein the instructions that, when executed, cause determining the subset of candidate content items comprise instructions that, when executed, cause:

determining a similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items.

20. The computer-readable medium of claim 19 , wherein the instructions that, when executed, cause determining the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items comprise instructions that, when executed, cause:

determining a cosine similarity distance between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items.

21. The computer-readable medium of claim 19 , wherein determining the subset of candidate content items further comprises:

comparing the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items with a threshold; and

determining that the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with the at least one candidate content item satisfies the threshold.

22. The computer-readable medium of claim 17 , wherein the instructions that, when executed, cause determining the at least one candidate content item comprise instructions that, when executed, cause:

determining a similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items.

23. The computer-readable medium of claim 22 , wherein the instructions that, when executed, cause determining the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items comprise instructions that, when executed, cause:

determining a cosine similarity distance between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items.

24. The computer-readable medium of claim 22 , wherein the instructions that, when executed, cause determining the at least one candidate content item further comprise instructions that, when executed, cause:

comparing the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items with a threshold; and

determining that the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with the at least one candidate content item satisfies the threshold.

25. A system comprising:

a device associated with a user; and

a computing device configured to:

determine a request associated with the user for a content recommendation;

determine, based on a viewing history associated with the user, at least one content item comprising video content;

determine, based on comparing encoded data associated with text associated with the at least one content item with encoded data associated with text associated with each candidate content item of a plurality of candidate content items, a subset of candidate content items of the plurality of candidate content items, wherein the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item each comprise one or more vectors configured to facilitate natural language processing;

determine, based on comparing encoded data associated with at least one image associated with the at least one content item with encoded data associated with at least one image associated with each of the subset of candidate content items, at least one candidate content item of the subset of candidate content items, wherein the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each of the subset of candidate content items each comprise one or more vectors configured to facilitate image classification or detection; and

send, to the device associated with the user, an indication of the at least one candidate content item.

26. The system of claim 25 , wherein the computing device is further configured to:

determine the encoded data associated with the text associated with each of the plurality of candidate content items using a first machine learning algorithm; and

determine the encoded data associated with the at least one image associated with each of the subset of candidate content items using a second machine learning algorithm.

27. The system of claim 25 , wherein the computing device is configured to determine the subset of candidate content items based on:

determining a similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items.

28. The system of claim 27 , wherein the computing device is configured to determine the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items based on:

determining a cosine similarity distance between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items.

29. The system of claim 27 , wherein the computing device is configured to determine the subset of candidate content items based on:

comparing the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with each candidate content item of the plurality of candidate content items with a threshold; and

determining that the similarity between the encoded data associated with the text associated with the at least one content item and the encoded data associated with the text associated with the at least one candidate content item satisfies the threshold.

30. The system of claim 25 , wherein the computing device is configured to determine the at least one candidate content item based on:

determining a similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items.

31. The system of claim 30 , wherein the computing device is configured to determine the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items based on:

determining a cosine similarity distance between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items.

32. The system of claim 30 , wherein the computing device is configured to determine the at least one candidate content item further based on:

comparing the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with each candidate content item of the subset of candidate content items with a threshold; and

determining that the similarity between the encoded data associated with the at least one image associated with the at least one content item and the encoded data associated with the at least one image associated with the at least one candidate content item satisfies the threshold.

Assignments (1)

Continuity (1)

References Cited (15)