Advertisement click rate prediction method

Abstract

The invention discloses an advertisement click rate prediction method. The method comprises the following steps: acquiring original instance data; preprocessing the original instance data; constructing an advertisement click rate prediction network model based on a convolutional neural network and an attention mechanism; training the advertisement click rate prediction network model; and testing the advertisement click rate prediction network model. The method solves a problem that the data is too sparse through employing the local feature extraction interaction capability and nonlinear capability of a convolutional neural network. In addition, the convolutional neural network is further utilized to perform feature interaction on the basis of the shallow interaction features to generate third-order or higher-order high-order features, so that the problem that an existing advertisement click rate prediction method is only limited to interaction of low-order features such as inner product calculation and outer product calculation is solved; meanwhile, by introducing an attention mechanism, useful features are further extracted on the basis of generating high-order features, and the influence of useless features on the network is reduced.

Description

technical field [0001] The invention relates to the technical field of Internet computing advertisements, in particular to a method for predicting the click-through rate of advertisements. Background technique [0002] With the widespread popularity of the Internet and the rapid development of big data technology, it has become possible for advertisers to use the Internet platform for advertising precision marketing. Compared with traditional advertising, online advertising has unique advantages in terms of coverage, flexibility and effect evaluation. One of the main goals of online advertising is to maximize an advertiser's revenue for a given budget, such as maximizing the number of clicks on an ad. Therefore, an important part of online advertising is to predict the click probability of a user who places an advertisement on an exposure opportunity, and the advertisement should be placed on an exposure opportunity with a high predicted click rate as much as possible. [...

AI Technical Summary

Problems solved by technology

[0004] In view of this, the present invention proposes an advertisement click-through rate prediction method, which can solve the difficulties in generating higher-order important interaction features and implicit interaction features, and only calculate the interaction of low-order features, and The technical problem of data being too sparse

Images