Ad fraud occurs when unscrupulous web or app publishers generate clicks, ad-impressions, or other events using bots that emulate human visitors to the web-page, app, or ads within. Ad fraud improperly diverts significant amounts of revenue to such publishers. Ad networks have long grappled with controlling or eliminating ad fraud.
This disclosure uses rewarded ads to issue challenges to willing users so as to discriminate between bots and humans. Users that are determined to be bots are tracked, ads appearing on associated web-pages or apps discarded, and such publishers noted for further action. The techniques enable detection of fraud using an API-approach, e.g., inside the ad, such that the use of a software development kit (SDK) for collecting ad-spam signals is lessened.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Toksoz, Tuna and Dukellis, John, "Rewarded tasks for reducing ad fraud", Technical Disclosure Commons, (December 03, 2018)