| CPC G06F 16/252 (2019.01) [G06F 16/24578 (2019.01); G06F 16/248 (2019.01); G06N 3/08 (2013.01); G06N 3/0895 (2023.01); G06N 3/09 (2023.01); G06N 3/091 (2023.01); G06N 3/092 (2023.01); G06N 3/094 (2023.01); G06N 3/096 (2023.01); G06N 3/098 (2023.01); G06N 3/0985 (2023.01)] | 20 Claims |
|
1. A computer-implemented method for ranking a plurality of content items for presentation to a user in response to a submission of a query, comprising:
generating, by at least one computer processor, a ranking score for each content item in the plurality of content items, wherein the generating the ranking score comprises, for each content item of the plurality of content items:
providing input to a deep machine learning (ML) model, the input including at least one or more query features associated with the query and one or more content item features associated with the content item;
mapping, by the deep ML model, each of the one or more query features to a plurality of embedding vectors using a corresponding plurality of hash functions, each of the embedding vectors being of a dimension;
processing, by a first multi-layer perceptron (MLP) of the deep ML model, a representation of the one or more content item features to generate processed content item features each as a vector of the dimension;
computing, by the deep ML model, dot products between all pairs of the embedding vectors and the processed content item features;
concatenating, by the deep ML model, the dot products with processed content item features to provide concatenated features;
post-processing, with a second MLP of the deep ML model, the concatenated features to provide post-processed concatenated features;
determining, by a third MLP of the deep ML model and based at least on the post-processed concatenated features, a first probability of a first type of user-item interaction between the user and the content item, wherein the first type of interaction between the user and the content item comprises the user launching the content item for playback;
determining, by a fourth MLP of the deep ML model and based at least on the post-processed concatenated features, a second probability of a second type of user-item interaction between the user and the content item, wherein the second type of interaction between the user and the content item comprises the user interacting with a user interface (UI) control associated with the content item to obtain information about the content item; and
calculating the ranking score for the content item based at least on the first probability and the second probability; and
ranking the plurality of content items for presentation to the user based on the ranking score associated with each content item in the plurality of content items.
|