Content Item Recommendations Based On Content Attribute Sequence

摘要

User created playlists can be analyzed to create a statistical language model indicating the likelihood that a particular sequence of content attributes will be found in a playlist created by a user, as well as the likelihood of any sequence of one or more content attributes following a playlist or partial playlist created by a user. The language model can be used to generate a recommended content attribute sequence based on a partial playlist of one or more content items. A recommended content item sequence that will be pleasant to a user when added to the partial playlist can be selected based on the recommended content attribute sequence.

主权项

1. A method comprising:
receiving, by a computer processor, data identifying a set of seed content items and a sequential order for the set of seed content items, wherein the set of seed content items includes at least one seed content item included in a first content item playlist, and the sequential order indicates an order in which seed content items from the set of seed content items are ordered to be performed in the first content item playlist; generating, by the computer processor, a seed content attribute sequence based on attributes of at least a first seed content item of the set of seed content items and the sequential order for the set of seed content items; generating, by the computer processor, a recommended content attribute sequence that is likely to follow the seed content attribute sequence, the recommended content attribute sequence determined from the seed content attribute sequence and a statistical language model based on an analysis of a set of reference content attribute sequences, wherein the set of reference content attribute sequences were generated from a set of user created content playlists; and selecting, by the computer processor, a set of recommended content items to be added to the first content item playlist and performed sequentially after the set of seed content items in a recommended sequential order, the set of recommended content item and the recommended sequential order selected based on content attributes of the set of recommended content items and the recommended content attribute sequence.