Detect fraudulent app installs
WHAT IT TESTS: signal design for install fraud. OUTLINE: click-to-install timing distributions, device and IP fingerprints, post-install engagement, and attribution anomalies.
WHAT IT TESTS: whether you can engineer signals that separate real users from install fraud like click injection, click flooding, and device farms. ANSWER OUTLINE: collect click-to-install time distributions, device and IP characteristics and clustering, post-install engagement depth and retention, and attribution anomalies such as a publisher with abnormally short click times or zero post-install activity; combine them into a model rather than one rule.
Read the original → interview
- #fraud-detection
- #mobile-attribution
- #cpi
- #feature-engineering
- #anti-fraud
Get five bites like this every day.
Tezvyn delivers a daily feed of 60-second tech bites with quizzes to lock in what you learn.