Website traffic trueness measurement method and system based on GAN and storage medium

Abstract

The invention discloses a GAN-based website traffic trueness measurement method and system and a storage medium. The method comprises the following steps: collecting website click event data; constructing a behavior sequence by using the website click event data; training the generative adversarial network model, continuously generating a simulated behavior sequence by using a generator in the generative adversarial network model, continuously learning how to judge whether the sequence is real or false by using a discriminator, and terminating the training until the loss of the generator and the discriminator is not reduced any more to obtain a trained generative adversarial network model; and using a discriminator of the trained generative adversarial network model to discriminate and measure the trueness of the constructed behavior sequence. The system comprises a data collection module, a behavior sequence construction module, a network model training module and a truth judgment module. According to the method, more accurate identification can be realized, the behavior rules of the user are learned by using a machine learning method, some rules which cannot be found manually can be found, and the method can adapt to different websites.

Description

technical field [0001] The invention belongs to the technical field of advertisement display effect monitoring, and in particular relates to a GAN-based website traffic authenticity measurement method, system and storage medium. Background technique [0002] Traffic cheating has become an open secret in the advertising industry, where there is a lot of non-human traffic (No-HumanTraffic), according to a new report from Virginia-based Distil Networks, a bot monitoring and API security expert, in 2018. More than a third of web traffic is driven by bots, and of that, more than 20% are driven by "bad bots" (as opposed to well-meaning bots such as search engine crawlers), a 6.4% increase from 2017. A bot is an automated software script / application used to perform repetitive tasks online. Common simulators in the domestic market include: NetEase MUMU simulator, Thunderbolt simulator, Xiaoyao simulator, Ye Shen simulator and Bluestack simulator, etc. These simulators can arbitrari...

AI Technical Summary

Problems solved by technology

[0004] 1) Once the rules are determined, they are rarely modified and cannot adapt to changes in the situation

[0005] 2) Rules rely on human experience and subjective judgment, and cannot be objective

[0006] 3) Different websites have different situations, and one-size-fits-all rules are difficult to meet the identification requirements

Images