Method and system for providing a structured topic drift for a displayed set of user comments on an article

1. A computer-implemented method of providing a structured topic drift for a displayed set of user comments on an article, the method comprising:

determining, using at least one processor, an ordered sequence of topical recommendations based on one or more properties of the displayed set of user comments and user characteristics using a sequential recommendation model, wherein each topical recommendation contains a plurality of user comments;

sampling, using the at least one processor, each of the topical recommendations in the ordered sequence to produce a first set of one or more user comments for each of the topical recommendations;

appending, using the at least one processor, one by one, the user comments from the first set of one or more user comments for each of the topical recommendations to bottom of the displayed set of user comments; and

updating, using the at least one processor, the sequential recommendation model based on a user response to the first set of one or more user comments for each of the topical recommendations;

determining, using at least one processor, a second ordered sequence of topical recommendations using the sequential recommendation model;

sampling, using the at least one processor, each of the topical recommendations in the second ordered sequence to produce a second set of one or more user comments for each of the topical recommendations; and

appending, using the at least one processor, one by one, the second set of one or more user comments for each of the topical recommendations to bottom of the displayed set of user comments.

2. The method as claimed in claim 1 , wherein the displayed set of user comments is associated with a topic.

3. The method as claimed in claim 1 , wherein determining the ordered sequence of topical recommendations further comprises

determining one or more features associated with the structured topic drift using a prediction function in the sequential recommendation model;

determining, using the sequential recommendation model, an initial internal state based on the one or more features, wherein the initial internal state is related to the user characteristics and is used to determine the ordered sequence of topical recommendations.

4. The method as claimed in claim 3 , wherein updating the sequential recommendation model based on the user response comprises

receiving a set of noisy observation input data, and determining, using the sequential recommendation model, a next internal state based on the noisy observation input data; and

wherein the next internal state is used to determine the second ordered sequence of topical recommendations.

5. The method as claimed in claim 3 , wherein the features associated with the structured topic drift comprise at least one of an effortless topic transition, a relevance to the displayed set of user comments to the user preferences, and a current determination of the structured topic drift.

6. The method as claimed in claim 1 , wherein the one or more user comments for each of the topical recommendations are sampled based on a ranking.

7. The method as claimed in claim 4 , wherein the noisy observation input data comprises at least one of a time spent viewing an individual user comments, a set of user preferences related to the user, or geographical information related to the user.

8. The method as claimed in claim 1 and further comprising

providing an option to a user to view a plurality of user comments similar to one or more of the displayed set of user comments and the one or more user comments for each of the topical recommendations.

9. A computer program product stored on a non-transitory computer-readable medium that when executed by a processor, performs a method of providing a structured topic drift for a displayed set of user comments on an article, comprising:

determining an ordered sequence of topical recommendations based on one or more properties of the displayed set of user comments and user characteristics using a sequential recommendation model;

sampling one or more user comments for a plurality of the topical recommendations in the ordered sequence to produce a set of one or more user comments;

appending, one by one, the one or more user comments for each of the topical recommendations to bottom of the displayed set of user comments; and

updating the sequential recommendation model based on a user response to the one or more user comments for each of the topical recommendations;

determining a second ordered sequence of topical recommendations using the sequential recommendation model;

sampling one or more user comments for a plurality of the topical recommendations in the second ordered sequence to produce a second set of one or more user comments; and

appending, one by one, the second set of one or more user comments for each of the topical recommendations to bottom of the displayed set of user comments.

10. The computer program product as claimed in claim 9 , wherein the displayed set of user comments is associated with a topic.

11. The computer program product as claimed in claim 9 , wherein determining the ordered sequence of topical recommendations comprises

determining one or more features associated with the structured topic drift using a prediction function in the sequential recommendation model;

determining, using the sequential recommendation model, an initial internal state based on the one or more features, wherein the initial internal state is used to determine the ordered sequence of topical recommendations.

12. The computer program product as claimed in claim 9 , wherein updating the sequential recommendation model based on the user response comprises

receiving a set of noisy observation input data, and determining, using the sequential recommendation model, a next internal state based on the noisy observation input data; and

wherein the next internal state is used to determine the second ordered sequence of topical recommendations.

13. The computer program product as claimed in claim 9 , wherein the topical recommendations are associated with different topics.

14. The computer program product as claimed in claim 9 , wherein the one or more user comments for each of the topical recommendations are sampled based on a ranking.

15. The computer program product as claimed in claim 9 , wherein the one or more user comments for each of the topical recommendations vary in number.

16. The computer program product as claimed in claim 9 and further comprising

providing an option to a user to view a plurality of user comments similar to one or more of the displayed set of user comments and the one or more user comments for each of the topical recommendations.

17. A system for providing a structured topic drift for a displayed set of user comments on an article, the system comprising:

one or more electronic devices;

a communication interface in electronic communication with the one or more electronic devices; a memory that stores instructions; and

a processor responsive to the instructions to

determine an ordered sequence of topical recommendations based on one or more properties of the displayed set of user comments and user characteristics using a sequential recommendation model;

sample one or more user comments for each of the topical recommendations to produce a set of one or more user comments for each of the topical recommendations;

append the one or more user comments for each of the topical recommendations to the displayed set of user comments; and

update the sequential recommendation model based on a user response to the one or more user comments for each of the topical recommendations.

18. The system as claimed in claim 17 and further comprising

an electronic storage device that stores the properties of the displayed set of user comments and the user characteristics, and the one or more user comments for each of the topical recommendations.

19. The system as claimed in claim 17 , wherein the processor is further responsive to the instructions to

determine a second ordered sequence of topical recommendations using the sequential recommendation model;

sample each of the topical recommendations in the second ordered sequence to produce a second set of one or more user comments for each of the topical recommendations; and

append the second set of one or more user comments for each of the topical recommendations to the displayed set of user comments.

20. The system as claimed in claim 17 , wherein the processor is further responsive to instructions to

determine one or more features associated with the structured topic drift using a prediction function in the sequential recommendation model;

determine, using the sequential recommendation model, an initial internal state based on the one or more features,

wherein the initial internal state is used to determine the ordered sequence of topical recommendations.

21. The system as claimed in claim 19 , wherein the processor is further responsive to instructions to

receive a set of noisy observation input data and determine, using the sequential recommendation model, a next internal state based on the noisy observation input data; and

wherein the next internal state is used to determine the second ordered sequence of topical recommendations.