Granted Patent B2

US 10,346,436 · App. 14/085,585 · Granted Jul 9, 2019

Abstract

Claims (68)

1. A computer implemented method of operation for a personalized content system, the method comprising:

obtaining electronic content over a computer network;

analyzing, by a processor, language of the electronic content to identify lexical features of the electronic content;

determining, by the processor, at least one topic of the electronic content based on the lexical features of the electronic content;

accessing, from storage device, an interest graph for a user that represents interests of the user in a plurality of topics, wherein each interest of the user is associated with an interest plot on the interest graph;

determining, by the processor, a velocity and an acceleration of each interest plot on the interest graph;

determining, by the processor, one or more future interests of the user associated with the interest graph based on the velocity and the acceleration of one or more interest plots on the interest graph;

determining, by the processor, a similarity between the electronic content and the interests of the user;

determining, by the processor, a second similarity between the electronic content and the future interests of the user by comparing the at least one topic of the electronic content and one or more future interests of the user on the interest graph; and

obtaining, over the computer network, personalized content that is personalized for the user based on the determined similarity and second similarity.

2. The method of claim 1 , wherein:

determining the similarity comprises determining the similarity between the interests of the user represented by the interest graph and the at least one topic of the electronic content; and

obtaining the personalized content is further based on the similarity between the interests of the user represented by the interest graph and the at least one topic of the electronic content.

3. The method of claim 2 , wherein:

determining the similarity between the interests of the user represented by the interest graph and the at least one topic of the electronic content comprises computing an interest score for the electronic content corresponding to the similarity between the interests of the user represented by the interest graph and the at least one topic of the electronic content based on the at least one topic of the electronic content and the interests of the user represented by the interest graph, and

obtaining the personalized content comprises obtaining the personalized content based on the interest score corresponding to the similarity between the interests of the user represented by the interest graph and the at least one topic of the electronic content.

4. The method of claim 1 , wherein determining the at least one topic of the electronic content based on the lexical features of the electronic content comprises:

comparing the lexical features of the electronic content to stored lexical features, wherein the stored lexical features are associated with one or more respective topics; and

determining the at least one topic of the electronic content responsive to comparing the lexical features of the electronic content to the stored lexical features.

5. The method of claim 1 , wherein:

determining the at least one topic of the electronic content comprises determining at least one confidence score corresponding to likelihood that the electronic content matches the at least one topic based on the lexical features of the electronic content, and

obtaining the personalized content comprises obtaining the personalized content based on the at least one confidence score and the interests of the user as represented by the interest graph.

6. The method of claim 1 , wherein the interest graph comprises a plurality of attachment levels representing the interests of user in the plurality of topics, and wherein obtaining the personalized content comprises:

obtaining the personalized content based on the at least one topic of the electronic content and the attachment levels representing the interests of user in the plurality of topics.

7. The method of claim 1 , further comprising:

generating a virality score corresponding to a popularity of the electronic content, wherein the personalized content is further obtained based on the virality score.

8. The method of claim 1 , further comprising:

predicting future interests of the user based on the interests of the user represented by the interest graph and historical interests of other users,

wherein obtaining the personalized content comprises obtaining the personalized content based on the at least one topic of the electronic content and the future interests of the user.

9. The method of claim 1 , further comprising:

obtaining more electronic content; analyzing language of the more electronic content to identify lexical features of the more electronic content; and

determining at least one topic of the more electronic content based on the lexical features of the more electronic content;

wherein the personalized content is further generated based on the at least one topic of the more electronic content.

10. The method of claim 1 , wherein obtaining the personalized content comprises:

generating a personalized content recommendation that is personalized for the user based on the at least one topic of the electronic content and the interests of the user represented by the interest graph.

11. A non-transitory computer-readable medium containing computer program instructions for a personalized content system, the instructions when executed by at least one processor cause the processor to:

obtain electronic content over a computer network;

access, from storage device, an interest graph for a user that represents interests of the user in a plurality of topics, wherein each interest of the user is associated with an interest plot on the interest graph;

determine, by the processor, a velocity and an acceleration of each interest plot on the interest graph;

determine, by the processor, one or more future interests of the user associated with the interest graph based on the velocity and the acceleration of one or more interest plots on the interest graph;

determine, by the processor, a similarity between the electronic content and the interests of the user;

determine, by the processor, a second similarity between the electronic content and the future interests of the user by comparing the at least one topic of the electronic content and one or more future interests of the user on the interest graph; and

obtain, over the computer network, personalized content that is personalized for the user based on the determined similarity and second similarity.

12. The computer-readable medium of claim 11 , wherein:

determining the similarity comprises determining a similarity between the interests of the user represented by the interest graph and the at least one topic of the electronic content; and

obtaining the personalized content is further based on the similarity between the interests of the user represented by the interest graph and the at least one topic of the electronic content.

13. The computer-readable medium of claim 12 , wherein:

the instructions to determine the similarity between the interests of the user represented by the interest graph and the at least one topic of the electronic content comprise instructions to compute an interest score for the electronic content corresponding to the similarity between the interests of the user represented by the interest graph and the at least one topic of the electronic content based on the at least one topic of the electronic content and the interests of the user represented by the interest graph, and

the instructions to obtain the personalized content comprise instructions to obtain the personalized content based on the interest score corresponding to the similarity between the interests of the user represented by the interest graph and the at least one topic of the electronic content.

14. The computer-readable medium of claim 11 , wherein the instructions to determine the at least one topic of the electronic content comprise instructions to:

compare the lexical features of the electronic content to stored lexical features, wherein the stored lexical features are associated with one or more respective topics; and

determine the at least one topic of the electronic content responsive to comparing the lexical features of the electronic content to the stored lexical features.

15. The computer-readable medium of claim 11 , wherein:

the instructions to determine the at least one topic of the electronic content comprise instructions to determine at least one confidence score corresponding to likelihood that the electronic content matches the at least one topic based on the lexical features of the electronic content, and

the instructions to obtain the personalized content comprise instructions to obtain the personalized content based on the at least one confidence score and the interests of the user as represented by the interest graph.

16. The computer-readable medium of claim 11 , wherein the interest graph comprises a plurality of attachment levels representing the interests of user in the plurality of topics, and wherein the instructions to obtain the personalized content comprise instructions to:

obtain the personalized content based on the at least one topic of the electronic content and the attachment levels representing the interests of user in the plurality of topics.

17. The computer-readable medium of claim 11 , the instructions further comprising instructions to cause the processor to:

generate a virality score corresponding to a popularity of the electronic content,

wherein the personalized content is further obtained based on the virality score.

18. The computer-readable medium of claim 11 , the instructions further comprising instructions to:

predict future interests of the user based on the interests of the user represented by the interest graph and historical interests of other users,

wherein the instructions to obtain the personalized content comprise instructions to obtain the personalized content based on the at least one topic of the electronic content and the future interests of the user.

19. The computer-readable medium of claim 11 , the instructions further comprising instructions to cause the processor to:

obtain more electronic content; analyze language of the more electronic content to identify lexical features of the more electronic content; and

determine at least one topic of the more electronic content based on the lexical features of the more electronic content; wherein the personalized content is further generated based on the at least one topic of the more electronic content.

20. The computer-readable medium of claim 11 , wherein the instructions to obtain the personalized content comprise instructions to:

generate a personalized content recommendation that is personalized for the user based on the one or more topics of the electronic content and the interests of the user represented by the interest graph.

Assignments (8)