Unveiling Mobile Ad Fraud Through FraudDetective Research
Explore the research on identifying and combating mobile ad fraud using FraudDetective, revealing insights on fraudulent activities in the mobile ad market and its impact on revenue.
Download Presentation
Please find below an Image/Link to download the presentation.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.
You are allowed to download the files provided on this website for personal or commercial use, subject to the condition that they are used lawfully. All files are the property of their respective owners.
The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author.
E N D
Presentation Transcript
Finding the Culprit Committing Mobile Ad Fraud Paper by Kim, Park, and Son Presented by Justin Kim
Problem In 2020, the US mobile ad market revenue was 187B USD How much of that was generated through fraudulent advertising?
Background Knowledge Ad service providers CPM, CPI, CPC Mobile fraud definition Generate fake, unwanted ad traffic, leading to fraudulent revenues Click fraud VS Impression fraud Click fraud generates fake clicks Impression fraud has ads in invisible elements
Motivation MAdLife, MAdFraud Cannot manifest casual relationships between user interaction and fraudulent activities Cannot pinpoint exact Android module which committed fraud Does not interact with target apps Both run on emulators Attempt to address these shortcomings with FraudDetective
FraudDetective 1. Crawl APKs from Google Play store, adding them to database 2. Task scheduler module creates tasks and schedules to available analysis worker 3. Analysis worker carries out dynamic testing, using Logcat to store information 4. Ad fraud detector verifies full stack trace for each task, reporting fraudulent activity
FraudDetective - Answer to Everything? 1. Cannot manifest casual relationships between user interaction and fraudulent activities a. FD is able to identify whether each trace originates from actual user interactions Cannot pinpoint exact Android module which committed fraud a. FD looks at the full stack trace of activities Does not interact with target apps a. FD separates user interaction from fraud detection policies Both run on emulators a. FD uses Android devices (Pixel 2) with modified Android Open Source Project 2. 3. 4.
FraudDetective - Findings Tested 48172 apps 34453 frauds found Type 1 - Forged user click Type 2 - No user interaction Type 3 - Other app invoked without interaction 98.6% of ad fraud is done by ad libraries
FraudDetective - Findings Out of 74 apps: 19 removed 49 updated No false negatives when compared to previous works
Closing Thoughts Google s restrictions towards ad fraud Possibilities of other fraudulent methods
