Data recommendation method, device and storage medium

Abstract

The invention discloses a data recommendation method, device, and storage medium. The method includes: acquiring a user's data recommendation request, and according to the data recommendation request, acquiring initial recommendation data corresponding to the user and characteristic data corresponding to the user; The request is converted into a request in a preset format, and the preset DeepFM model is obtained from the model server according to the request in the preset format, wherein the request in the preset format meets the identification requirements of the model server; according to the preset DeepFM model and feature data, generate Update data; generate target recommendation data based on updated data and initial recommendation data; push target recommendation data to users. Based on the preset DeepFM model, the solution generates update data, and updates the initial recommendation data according to the update data to generate target recommendation data, which improves the efficiency of data recommendation.

Description

technical field [0001] The present invention relates to the field of communication technology, in particular to a data recommendation method, device and storage medium. Background technique [0002] With the popularization of mobile terminals such as smart phones and tablet computers, users can obtain more and more information through the terminals. In order to enable the user to quickly obtain the desired information, information recommendations can be made according to the user's hobbies. [0003] Existing recommender systems generally use a logistic regression (LR) algorithm. The LR algorithm has the advantages of low complexity and easy understanding and implementation, but it also has limitations. The LR algorithm can only automatically describe low-order features, and manual feature combination is required in the process of combining high-order features. However, the workload in the process of feature combination is large, resulting in low recommendation efficiency. ...

AI Technical Summary

Problems solved by technology

[0003] The existing recommendation system generally uses the Logistic Regression (LR) algorithm. The LR algorithm has the advantages of low complexity, easy understanding and implementation, but it also has limitations.

The LR algorithm can only automatically describe low-order features. In the process of combining high-order features, manual feature combination is required.

However, the workload in the feature combination process is relatively large, resulting in low recommendation efficiency.

Images