Click fraud detection on the advertiser side

Haitao Xu, Daiping Liu, Aaron Koehl, Haining Wang, Angelos Stavrou

Research output: Chapter in Book/Report/Conference proceedingConference contribution

29 Scopus citations

Abstract

Click fraud - malicious clicks at the expense of pay-per-click advertisers - is posing a serious threat to the Internet economy. Although click fraud has attracted much attention from the security community, as the direct victims of click fraud, advertisers still lack effective defense to detect click fraud independently. In this paper, we propose a novel approach for advertisers to detect click frauds and evaluate the return on investment (ROI) of their ad campaigns without the helps from ad networks or publishers. Our key idea is to proactively test if visiting clients are full-fledged modern browsers and passively scrutinize user engagement. In particular, we introduce a new functionality test and develop an extensive characterization of user engagement. Our detection can significantly raise the bar for committing click fraud and is transparent to users. Moreover, our approach requires little effort to be deployed at the advertiser side. To validate the effectiveness of our approach, we implement a prototype and deploy it on a large production website; and then we run 10-day ad campaigns for the website on a major ad network. The experimental results show that our proposed defense is effective in identifying both clickbots and human clickers, while incurring negligible overhead at both the server and client sides.

Original languageEnglish (US)
Title of host publicationComputer Security, ESORICS 2014 - 19th European Symposium on Research in Computer Security, Proceedings
PublisherSpringer Verlag
Pages419-438
Number of pages20
EditionPART 2
ISBN (Print)9783319112114
DOIs
StatePublished - 2014
Externally publishedYes
Event19th European Symposium on Research in Computer Security, ESORICS 2014 - Wroclaw, Poland
Duration: Sep 7 2014Sep 11 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 2
Volume8713 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference19th European Symposium on Research in Computer Security, ESORICS 2014
Country/TerritoryPoland
CityWroclaw
Period9/7/149/11/14

Keywords

  • Click Fraud
  • Feature Detection
  • Online Advertising

ASJC Scopus subject areas

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Click fraud detection on the advertiser side'. Together they form a unique fingerprint.

Cite this