Methods and systems for generating and presenting content recommendations for new users
Systems and methods for generating and presenting content recommendations to new users during or immediately after the onboarding process, before any history of the new user's viewed content is available. A machine learning or other model may be trained to determine clusters of content genre values corresponding to genres of content watched by viewers. Clusters are thus associated with popular groupings of content genres viewed by many users. Clusters representing popular groupings of content genres may be selected for new users, and content corresponding to the selected clusters may be recommended to the new users as part of their onboarding process. A sufficient amount of content may be selected to fully populate any content recommendation portion of a new user onboarding page.
1 . A computing device implemented method comprising:
accessing a collection of user preference indication data based at least in part on stored preference indications for content items of a first plurality of users;
identifying a second plurality of users from the first plurality of users, wherein each user of the second plurality of users has a stored preference indication for a greater than a predetermined threshold number of content items;
training a machine learning model to identify clusters of user preference indication data in the collection of user preference indication data, wherein the preference indication data of the second plurality of users is unused in the collection of user preference indication data for the training; and
testing the machine learning model using the preference indication data of the second plurality of users;
receiving an indication of one or more content interactions of a first user;
determining content preferences of the first user based at least in part on the indication of one or more content interactions of the first user;
identifying, using the machine learning model after training and testing, a content item to be recommended based at least in part on the determined content preferences of the first user; and
providing an indication of the identified content item.
2 . The method of claim 1 , wherein the identifying the content item to be recommended is made for a new user as part of an onboarding process of the new user; and the providing of the indication of the identified content item comprises:
generating of display of the indication of the identified content item.
3 . The method of claim 1 , further comprising:
determining a set of user classes to which a first user belongs by selecting one or more clusters of user preference indication data for media content from among a plurality of clusters of user preference data for media content, each cluster representing content preference indications of a subset of the first plurality of users, the plurality of users excluding the first user.
4 . The method of claim 1 , wherein the machine learning model comprises one or more of an expectation maximization (EM) model, a k-means model, or a k-nearest neighbor model.
5 . The method of claim 1 , wherein the second plurality of users comprises a set of user content preference indication data of users with greater than a threshold frequency of content consumption.
6 . The method of claim 1 , wherein the indication of the identified content item is provided to a first user, the method further comprising:
selecting, using the machine learning model after training and testing, a second set of content recommendations according to the determined content preferences of the first user; and
generating for display indications of the second set of content recommendations.
7 . The method of claim 1 , wherein the providing the indication of the identified content item is caused to be displayed as a first content recommendations page displaying a plurality of indications of content items.
8 . The method of claim 1 , wherein each cluster of the clusters of user preference indication data in the collection of user preference indication data corresponds to preferences for a content genre.
9 . The method of claim 1 , wherein the identifying the content item to be recommended is made for a new user as part of an onboarding process of the new user; and
wherein the identifying, using the machine learning model after training and testing, the content item to be recommended by weighting less the new user's content preferences for content items.
10 . A system comprising:
communication circuitry configured to:
access a collection of user preference indication data based at least in part on stored preference indications for content items of a first plurality of users; and
processing circuitry configured to:
identify a second plurality of users from the first plurality of users, wherein each user of the second plurality of users has a stored preference indication for a greater than a predetermined threshold number of content items;
train a machine learning model to identify clusters of user preference indication data in the collection of user preference indication data, wherein the preference indication data of the second plurality of users is unused in the collection of user preference indication data for the training;
test the machine learning model using the preference indication data of the second plurality of users;
receive an indication of one or more content interactions of a first user;
determine content preferences of the first user based at least in part on the indication of one or more content interactions of the first user; and
identify, using the machine learning model after training and testing, a content item to be recommended based at least in part on the determined content preferences of the first user; and
the communication circuitry further configured to:
provide an indication of the identified content item.
11 . The system of claim 10 , wherein the identifying, by the processing circuitry, the content item to be recommended is made for a new user as part of an onboarding process of the new user; and the providing, by the communication circuitry, of the indication of the identified content item comprises:
generating of display of the indication of the identified content item.
12 . The system of claim 10 , wherein the system is further configured
to determine a set of user classes to which a first user belongs by selecting one or more clusters of user preference indication data for media content from among a plurality of clusters of user preference data for media content, each cluster representing content preference indications of a subset of the first plurality of users, the plurality of users excluding the first user.
13 . The system of claim 10 , wherein the machine learning model comprises one or more of an expectation maximization (EM) model, a k-means model, or a k-nearest neighbor model.
14 . The system of claim 10 , wherein the second plurality of users comprises a set of user content preference indication data of users with greater than a threshold frequency of content consumption.
15 . The system of claim 10 , wherein the indication of the identified content item is provided to a first user, and wherein the system processing circuitry is further configured to:
select, using the machine learning model after training and testing, a second set of content recommendations according to the determined content preferences of the first user; and
instruct to generate for display indications of the second set of content recommendations.
16 . The system of claim 10 , wherein the providing, by the communication circuitry, the indication of the identified content item is caused to be displayed as a first content recommendations page displaying a plurality of indications of content items.
17 . The system of claim 10 , wherein each cluster of the clusters of user preference indication data in the collection of user preference indication data corresponds to preferences for a content genre.
18 . The system of claim 10 , wherein the identifying, by the processing circuitry, the content item to be recommended is made for a new user as part of an onboarding process of the new user; and
wherein the identifying, using machine learning model after training and testing, the content item to be recommended by weighting less the new user's content preferences for content items.