Machine-learned recommender system for performance optimization of network-transferred electronic content items

A method comprising:

• generating a prediction model based on training data that comprises a plurality of training instances, each of which comprises a feature value for each of a plurality of features and a label of a target variable;
• wherein the training data is related to multiple campaign-level or serving attributes of each campaign of a plurality of content delivery campaigns;
• receiving a first plurality of feature values, for the plurality of features, of a first content delivery campaign that comprises one or more content items that are separate from the first content delivery campaign;
• inputting, into the prediction model, the first plurality of feature values to generate output;
• based on the output from the prediction model, identifying one or more recommendations;
• causing the one or more recommendations to be presented on a computing device associated with the first content delivery campaign;
• wherein the method is performed by one or more computing devices.

One or more storage media storing instructions which, when executed by one or more processors, cause:

• generating a decision tree based on training data that comprises a plurality of training instances, each of which comprises a feature value for each of a plurality of features and a label of a target variable;
• wherein the plurality of features correspond to campaign-level or serving attributes of multiple content delivery campaigns;
• wherein the decision tree comprises one or more subtrees;
• receiving a first plurality of feature values, for the plurality of features, of a first content delivery campaign that comprises one or more content items that are separate from the first content delivery campaign;
• traversing the decision tree using the first plurality of feature values to generate output;
• based on the output, identifying one or more recommendations;
• causing the one or more recommendations to be presented on a computing device associated with the first content delivery campaign.