Click rate estimation method based on multi-domain partition integrated network

Abstract

The invention discloses a click rate estimation method based on a multi-domain partition integrated network, which is characterized by adopting three parallel modules, providing two multi-domain partition strategies for original onehot form features and dividing into region vectors; independently embedding each area vector by using a segmentation embedding method to obtain an embedded layer vector; and sharing the embedded vector to a mining middle-order interactive network and a high-order interactive neural network. The middle-order interactive network adopts the FFM to extract the feature interaction between the regions, and the high-order interactive part introduces the integration thought on the basis of the neural network, so that the degree of parallelism of the network is widened, and the expression ability of positive features is enhanced. According to the method, independent features and interactive features are considered at the same time, the expression ability of an embedded layer is enriched by using the idea of segmented embedding, the neural network is extended under the condition that space complexity is not introduced, the problem of gradient disappearance is effectively solved, and the click estimation ability is improved.

Description

technical field [0001] The invention relates to a method for estimating a click rate, in particular to a method for estimating an advertisement click rate, and belongs to the field of advertisement recommendation. Background technique [0002] With the development of the Internet for decades, computational advertising is undoubtedly the first choice for all traffic monetization, and it is also the main source of income for most Internet companies. When a user clicks on an advertisement on the advertising platform, the provider of the advertisement, that is, the advertiser, will pay a certain fee to the platform. Therefore, the revenue of advertising platforms depends on their ability to accurately predict the ads that users may click and display them to users. They need to use efficient advertising recommendation technology to help advertisers find potential advertising consumers and achieve mutual benefits. [0003] In the advertising click-through rate prediction task, in...

AI Technical Summary

Problems solved by technology

[0004] However, these methods ignore the information differences between different feature domains, lack corresponding considerations for mid-level feature combinations, and lack the extension of the depth and width of the neural network. Based on this, the estimation method for the click-through rate of advertisements and technology requires further innovation

Images