Digital content recommendations based on user comments

1. A method comprising:

receiving, by a processor, comments associated with first digital content, wherein each of the comments are submitted by a first user for a corresponding portion of the first digital content;

determining, by the processor, opinions attributed to the first user with respect to each of the corresponding portions of the first digital content based on the comments, wherein the attributed opinions correspond to preference levels for each corresponding portion of the first digital content, and wherein determining the attributed opinions includes:

parsing the comments to identify a term included in the comments, and

determining the attributed opinions based on the identified term;

generating, by the processor, a first preference profile including a preference level associated with each corresponding portion of the first digital content, wherein generating the first preference profile includes:

identifying a series of data points that correspond to the preference levels,

performing interpolation to generate a curve that connects the data points, and

performing extrapolation to fit the curve to one or more other portions of the first digital content for which no comment is received from the first user;

comparing, by the processor, the first preference profile to a plurality of second preference profiles associated with the first digital content for a plurality of second users;

clustering, by the processor, the first user and at least one of the plurality of second users, wherein the second preference profile for the at least one of the plurality of second users differs from the first preference profile by less than a threshold amount;

identifying, by the processor, second digital content liked by the at least one of the plurality of second users which has not been accessed by the first user; and

providing, by the processor and for presentation to the first user, a recommendation identifying the second digital content.

2. The method of claim 1 , wherein determining the opinions attributed to the first user includes:

determining the corresponding portions of the first digital content that are displayed when the comments are received, and

determining the attributed opinions with respect to the corresponding portions of the first digital content.

3. The method of claim 2 , wherein determining the corresponding portions of the first digital content includes:

determining times when the comments are received; and

determining the corresponding portions of the first digital content based on the times.

4. The method of claim 2 , wherein one or more second users, of the plurality of second users, have a similar opinion as the first user with respect to at least one of the corresponding portions of the first digital content.

5. The method of claim 2 , wherein identifying the second digital content includes:

identifying a plurality of candidate digital content associated with the at least one of the plurality of second users which has not been accessed by the first user; and

selecting the second digital content from the plurality of candidate digital content based on a comparison of the attributed opinions with respect to the corresponding portions of the first digital content and respective opinions of the at least one of the plurality of second users with respect to the corresponding portions of the first digital content.

6. The method of claim 1 , wherein identifying the second digital content includes:

identifying a plurality of candidate digital content liked by the at least one of the plurality of second users which has not been accessed by the first user; and

selecting the second digital content from the plurality of candidate digital content based on a respective quantity of the at least one of the plurality of second users that like each of the plurality of candidate digital content.

7. The method of claim 1 , wherein each of the corresponding portions of the first digital content spans a timeframe of a same length of time.

8. A device comprising:

a memory configured to store comments associated with first digital content, wherein each of the comments are submitted by a first user for a corresponding portion of the first digital content;

and

a processor configured to:

parse the comments to identify a term included in the comments,

determine opinions attributed to the first user with respect to each of the corresponding portions of the first digital content based on each of the terms,

generate a first preference profile including a preference level associated with each corresponding portion of the first digital content, wherein the processor, when generating the first preference profile, is configured to:

identify a series of data points that correspond to the preference levels,

perform interpolation to generate a curve that connects the data points, and

perform extrapolation to fit the curve to one or more other portions of the first digital content for which no comment is received from the first user,

compare the first preference profile to a plurality of second preference profiles associated with the first digital content for a plurality of second users,

cluster the first user and at least one of the plurality of second users, wherein the second preference profile for the at least one of the plurality of second users differ from the first preference profile by less than a threshold amount,

identify second digital content accessed by the at least one of the plurality of second users which has not been accessed by the first user, and

provide, for presentation to the first user, a recommendation identifying the second digital content.

9. The device of claim 8 , wherein the processor, when determining the opinions attributed to the first user, is further configured to:

identify the corresponding portions of the first digital content that are displayed when the comments are received, and

determine the attributed opinions with respect to the corresponding portions of the first digital content.

10. The device of claim 9 , wherein the processor, when determining the corresponding portions of the first digital content, is further configured to:

determine times when comments are received, and

determine the corresponding portions of the first digital content based on the times.

11. The device of claim 9 , wherein one or more second users, of the plurality of second users, have a similar opinion as the first user with respect to at least one of the corresponding portions of the first digital content.

12. The device of claim 9 , wherein the processor, when identifying the second digital content, is further configured to:

identify a plurality of candidate digital content associated with the at least one of the plurality of second users which has not been accessed by the first user, and

select the second digital content from the plurality of candidate digital content based on a comparison of the attributed opinions with respect to the corresponding portions of the first digital content and respective opinions of the at least one of the plurality of second users with respect to the corresponding portions of the first digital content.

13. The device of claim 8 , wherein the processor, when identifying the second digital content, is further configured to:

identify a plurality of candidate digital content liked by the at least one of the plurality of second users which has not been accessed by the first user, and

select the second digital content from the plurality of candidate digital content based on a respective quantity of the at least one of the plurality of second users that like each of the plurality of candidate digital content.

14. The device of claim 8 , wherein each of the corresponding portions of the first digital content is a timeframe of a same length of time.

15. A non-transitory computer-readable medium to store instructions comprising:

one or more instructions that, when executed by a processor, cause the processor to:

identify comments associated with first digital content, wherein each of the comments are submitted by a first user for a corresponding portion of the first digital content,

parse the comments to determine an opinion attributed to the first user with respect to each of the corresponding portions of the first digital content,

generate a first preference profile including a preference level associated with each corresponding portion of the first digital content, wherein when generating the first preference profile, the processor to:

identify a series of data points that correspond to the preference levels,

perform interpolation to generate a curve that connects the data points, and

perform extrapolation to fit the curve to one or more other portions of the first digital content for which no comment is received from the first user,

compare the first preference profile to a plurality of second preference profiles associated with the first digital content for a plurality of second users,

cluster the first user and at least one of the plurality of second users, wherein the second preference profile for the at least one of the plurality of second users differs from the first preference profile by less than a threshold amount,

identify second digital content accessed by the at least one of the plurality of second users which has not been accessed by the first user, and

provide, for presentation to the first user, a recommendation identifying the second digital content.

16. The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions to determine the opinions attributed to the first user further include:

one or more instructions to:

determine the corresponding portions of the first digital content that are displayed when the comments are received, and

determine the attributed opinions with respect to the corresponding portions of the first digital content.

17. The non-transitory computer-readable medium of claim 16 , wherein the one or more instructions to determine the corresponding portions of the first digital content further include:

one or more instructions to:

determine times when the comments are received, and

determine the corresponding portions of the first digital content based on the times.

18. The non-transitory computer-readable medium of claim 16 , wherein the one or more instructions to identify the second digital content further include:

one or more instructions to:

identify a plurality of candidate digital content associated with the at least one of the plurality of second users which has not been accessed by the first user, and

select the second digital content from the plurality of candidate digital content based on a comparison of the attributed opinions with respect to the corresponding portions of the first digital content and respective opinions of the at least one of the plurality of second users with respect to the corresponding portions of the first digital content.

19. The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions to identify the second digital content further include:

one or more instructions to:

identify a plurality of candidate digital liked by the at least one of the plurality of second users which has not been accessed by the first user, and

select the second digital content from the plurality of candidate digital content based on a respective quantity of the at least one of the plurality of second users that like each of the plurality of candidate digital content.

20. The non-transitory computer-readable medium of claim 15 , wherein each of the corresponding portions of the first digital content is a timeframe of a same length of time.