Method, device, server, and storage medium for estimating advertisement click rate

Abstract

The invention relates to an advertisement click rate estimation method and device, a server and a storage medium, and the method comprises the steps: training a first network model according to firstsample data, wherein the first network model is used for estimating the advertisement click rate; Respectively creating a first loss function between the output result of the trained first network model and second sample data, and a second loss function between the output result of the trained first network model and the output result of the trained second network model, wherein the second networkmodel is used for estimating the advertisement click rate; Training the second network model according to the first loss function and the second loss function; And estimating the advertisement clickrate to be estimated by using the trained second network model. According to the invention, the estimation accuracy of the advertisement click rate is improved, and the calculation amount for trainingthe second network model is also reduced.

Description

technical field [0001] The present disclosure relates to the field of computer technology, and in particular to a method, device, server and storage medium for estimating the click-through rate of advertisements. Background technique [0002] Advertisement click-through rate estimation of Internet advertisements is very important, and it is the core of advertisement delivery mechanism and strategy. The advertising click-through rate estimation model is to model the advertising click-through rate based on the user's behavior data on the advertisement. [0003] With the development of deep learning technology, deep learning models have become the mainstream of advertising click rate prediction models. Complex deep learning models (such as convolutional neural networks and recurrent neural networks) can more accurately predict the click-through rate of advertisements, but the amount of calculation required is also very large. Simple deep learning models (such as multi-layer p...

AI Technical Summary

Problems solved by technology

Complex deep learning models (such as convolutional neural networks and recurrent neural networks) can more accurately predict the click-through rate of advertisements, but the amount of calculation required is also very large

Simple deep learning models (such as multi-layer perceptrons) require less computation, but are less accurate than complex deep learning models

[0004] In related technologies, the advertising click-through rate estimation model is often only trained by a simple deep learning model, and the accuracy of the advertising click-through rate estimation is low, or it is only trained by a complex deep learning model, and the training calculation is large

Images