User-specific media playlists
A media recommendation system may score media items according to user recommendations, popularity, and/or recency. The scores may be weighted to produce an overall score for each media item. Media items may be added to a pool for a specific user, from which media items are selected for playback. The contents of the pool may be modified based upon user feedback and other data. The pool may be modified dynamically and/or in real time as media items are consumed or rated by the user.
1. A method comprising:
receiving, at a processor, a feedback signal about a first media item that is selected to be played from a first pool of media items, wherein the first pool of media items is sorted based on distances between vector representations of the media items; and
modifying, at the processor and based upon the feedback signal, contents of the first pool of media items to form a second pool of media items by changing the selection of the media items in the first pool to include at least a second media item that is not in the first pool, and sorting the media items in the second pool based upon a distance between a vector representation of the first media item and a vector representation of the second media item.
2. The method of claim 1 , further comprising:
modifying, at the processor, the contents of the second pool as one or more of the media items in the second pool are selected to be played, wherein the modifying is based on the selection of the one or more media items from the second pool.
3. The method of claim 1 , further comprising:
modifying, at the processor, the contents of the second pool as one or more of the media items in the second pool are selected to be played, wherein the modifying is based on the selection of a rating of the one or more media items in the second pool.
4. The method of claim 1 , wherein at least one of the first pool and the second pool comprises a subset of a plurality of media items available within a computerized media recommendation system.
5. The method of claim 1 , further comprising:
generating, at the processor, the vector representations of the media items in at least one of the first pool and the second pool.
6. The method of claim 1 , further comprising:
generating, at the processor, the first pool from a plurality of the media items that are available within a computerized media recommendation system.
7. The method of claim 1 , wherein the generating the first pool comprises:
generating, at the processor, the first pool based on a seed media item.
8. The method of claim 1 , further comprising:
selecting, at the processor, the first media item as a seed media item.
9. The method of claim 8 , wherein the seed media item is of a type selected from the group consisting of: trending, emerging, recent, liked, selected, active, and recommended.
10. The method of claim 8 , wherein media item is defined by at least one selected from a group consisting of: a count over a specific duration of time of a consumption of a media item, and a count over the specific duration of time of receipts of positive feedback about the media item.
11. The method of claim 10 , wherein the media item in the count of the consumption of the media item comprises at least one selected from the group consisting of: a media item determined to be less well-known than an average well-known media item, a media item consumed by a user from a source external to a computerized media recommendation system, and a media item being consumed by users connected to the user by a social media network.
12. The method of claim 10 , wherein the media item in the count of the receipts of the positive feedback comprises at least one selected from a group consisting of: a media item associated with an affirmative action to express the positive feedback, and a media item selected by an individual associated with an entity that controls a computerized media recommendation system.
13. The method of claim 8 , wherein the selecting of the seed media item is based on a profile vector of a user.
14. The method of claim 13 , further comprising:
generating, at the processor, the profile vector of the user by normalizing a sum of vector representations of media items previously consumed by the user.
15. The method of claim 14 , further comprising:
updating, at the processor, the profile vector based on one or more selections by the user.