Title:
|
TIME AND SPACE CONTEXTUAL INFORMATION IMPROVES CLICK QUALITY ESTIMATION |
Author(s):
|
Mehmed Kantardzic , Brent Wenerstrom , Chamila Walgampaya , Oleksandr Lozitskiy , Sean Higgins , Darren King |
ISBN:
|
978-972-8924-89-8 |
Editors:
|
Sandeep Krishnamurthy |
Year:
|
2009 |
Edition:
|
Single |
Keywords:
|
Click Fraud, Click Fraud Prevention, Click Traffic Quality, Web Analytics |
Type:
|
Full Paper |
First Page:
|
123 |
Last Page:
|
130 |
Language:
|
English |
Cover:
|
|
Full Contents:
|
click to dowload
|
Paper Abstract:
|
Click fraud is a type of internet crime that occurs in pay per click online advertising when a person, automated script, or
computer program imitates a legitimate user clicking on an ad, for the purpose of generating a charge per click without
having actual interest in the adÂ’s content. We propose Collaborative Click Fraud Detection and Prevention system
(CCFDP) V1.0, which integrates client and server side data to score the quality of incoming clicks. In this paper we detail
the outlier detection module which describes clicks in terms of space and time context. This module compares past data
with the current context as a preprocessing step. Our system will then combine the additional time and space
characteristics with the characteristics of a click to score the quality of incoming clicks. We believe that no other
commercial or research system for click fraud detection analyze comprehensively time and space context of each click
for better estimation of click traffic quality. Some commercial solutions give only partial solutions expressed through
their rules and triggers. We found 34.6% of clicks in an application to real data had outlying attribute-values in time and
space. |
|
|
|
|