Advertisement promotion method combined with advertiser effect feedbacks

Abstract

The invention discloses an advertisement promotion method combined with advertiser effect feedbacks. The method comprises: according to advertisement effect fed back by an advertiser, all users to whom an advertisement is pushed are divided into accurate users and other users; a set of the accurate users is taken as a positive user sample and a set of the other users is a negative user sample, user features are extracted from the positive user sample and the negative user sample, the extracted user features are processed, and a prediction model is obtained; for an unknown user, user features of the unknown user are extracted and applied to the prediction model, and the probability of the unknown user becoming an accurate user is obtained; the advertisement is pushed to the unknown user whose probability of becoming an accurate user reaches a preset value, and is not pushed to the unknown user whose probability of becoming an accurate user does not reach the preset value. The method can improve the cost-performance ratio of advertisement delivery and the promotion efficiency of the whole advertisement platform.

Description

technical field [0001] The invention relates to the technical field of advertising promotion, in particular to an advertising promotion method combined with advertiser effect feedback. Background technique [0002] Online advertising has become the mainstream form of the advertising industry. According to the effect, online advertisement can be divided into brand display advertisement and effect-based billing advertisement. Among them, the purpose of brand display advertising is to let target users gain brand awareness through display exposure, such as traditional TV advertisements, video placement ads are all this type of advertisement; the purpose of performance-based billing advertising is to let target users recognize the brand through display exposure And generate operational behavior to obtain income, such as user clicks. [0003] Correspondingly, the current mainstream technical solutions for online advertising include user orientation, keyword matching or a combina...

AI Technical Summary

Problems solved by technology

[0004] The main purpose of advertising is to help advertisers acquire users. However, the disadvantage of the above two advertising methods is that they cannot accurately acquire users.

There are two reasons for this: First, advertisers cannot define precise users. Whether it is user targeting or keyword matching, it is more based on prior experience and knowledge. For advertisers, this is very dependent on advertising operations The quality and experience of personnel; secondly, the online advertising platform cannot define accurate users, and the online advertising platform can approximately define accurate users through page stay time, user clicks and other behaviors, but there is still a deviation from the actual accurate users

Images