Detection of Click Spamming in Mobile Advertising

Abstract

Most of the marketing expenditures in mobile advertising are conducted through real-time bidding (RTB) marketplaces, in which ad spaces of the sellers (publishers) are auctioned for the impression of the buyers’ (advertisers) mobile apps. One of the most popular pricing models in RTB marketplaces is cost-per-install (CPI). In a CPI campaign, publishers place mobile ads of the highest bidders in their mobile apps and are paid by advertisers only if the advertised app is installed by a user. CPI pricing model causes some publishers to conduct an infamous fraudulent activity, known as click spamming. A click spamming publisher executes clicks for lots of users who have not authentically made them. If one of these users hears about the advertised app organically (say, via TV commercial) afterwards and installs it, this install will be attributed to the click spamming publisher. In this study, we propose a novel multiple testing procedure which can identify click spamming activities using the data of click-to-install time (CTIT), the time difference between the click of a mobile app’s ad and the first launch of the app after the install. We statistically show that our procedure has a false-positive error rate of 5% in the worst case. Finally, we run an experiment with 30 publishers, half of which are fraudulent. According to the results of the experiment, all non-fraudulent publishers are correctly identified and 73% of the fraudulent publishers are successfully detected.