27th International Conference on Distributed Computing Systems (ICDCS '07) 2007
DOI: 10.1109/icdcs.2007.124
|View full text |Cite
|
Sign up to set email alerts
|

On Hit Inflation Techniques and Detection in Streams of Web Advertising Networks

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
26
0

Year Published

2008
2008
2021
2021

Publication Types

Select...
4
2
1

Relationship

1
6

Authors

Journals

citations
Cited by 37 publications
(26 citation statements)
references
References 16 publications
0
26
0
Order By: Relevance
“…Click fraud on a larger scale uses a script to simulate human clicking or make visitors automatically click on ads [143][144][145][146][147][148][149][150][151][152]. Botnets can also be used for click fraud [153].…”
Section: How It Workmentioning
confidence: 99%
See 1 more Smart Citation
“…Click fraud on a larger scale uses a script to simulate human clicking or make visitors automatically click on ads [143][144][145][146][147][148][149][150][151][152]. Botnets can also be used for click fraud [153].…”
Section: How It Workmentioning
confidence: 99%
“…It cannot be detected by the above method, since the clicks are from many different visitors. It is necessary to monitor the click-through rate (CTR), which is the number of clicks divided by that of impressions (i.e., ad views) [147,150]. It is an estimate of the probability in the ad network that an impression leads to a click.…”
Section: Forms Of Click Fraud and Technical Responsesmentioning
confidence: 99%
“…Click fraud schemes [23] have the dual effect of inflating the number of "click-throughs" on the Ads posted on a website and increasing the popularity of the website as the number of hits increases. While click-fraud may directly affect the ranks produced by direct counting schemes (e.g., those of Alexa), its effect on our ranks is more limited.…”
Section: Resilience To Fraudulent Inflationmentioning
confidence: 99%
“…However, since the type of sanitization that is applied, as well as the specific data collected, are completely at the discretion of the data collector, large cross-sections of the population shy away from deploying these agents. Moreover, the mere clientside nature of these collection schemes opens the door to abuse (be it via click fraud [23] or other manipulation recipes [1]) that directly affect the integrity of the results. Recently, the prevalence of these fraudulent behaviors has raised so much doubts about the integrity and authenticity of these ranking measures that it captured the attention of the mainstream press (e.g., [22]).…”
Section: Introductionmentioning
confidence: 99%
“…Multi stream analysis arises in many contexts, including large scale web advertising networks, the internet, sensor networks, and real time data analysis [6]. In large scale web-based advertising networks, clicks originating from different internet hosts result in multiple click streams that need to be merged at a single server for analysis and identification of suspicious publishers or IP addresses [21]. In the Internet, network monitors at core routers process streams of packets arriving from different origins for accounting purposes, and detection of malicious activities such as Distributed Denial of Service attacks [29].…”
Section: Introductionmentioning
confidence: 99%