Degree Name

Master of Research


School of Information and Computer Science


The thesis mainly focuses on applying fuzzy and rough set theory to the mobile adver-tisement fraud detection. We first present an application of fuzzy set theory to implement mobile advertisement anti-fraud systems. One of the main problems of mobile anti-fraud is the lack of evidence to prove a user to be a fraudster. In this thesis, we implement an application of fuzzy set theory to demonstrate how to detect cheaters. The advantage of this method is such that the hardship in detecting fraudsters in small data samples can be avoided. We achieve this by giving each user a suspicious degree showing how likely the user is cheating and decide whether a group of users (like all users of a certain APP) together could be fraudsters according to the average suspicious degree. This makes the process more accurate as the data of a single user is too small to be predictable. We also develop another method with the application of rough set theory. The advantage of the second anti-fraud method proposed in this work is that the method is hard to counter. It means that avoiding the detection of this method is very difficult for fraudsters. Ever since smartphone and mobile internet became popular, mobile advertisement fraud and anti-fraud have become two competitors both trying to suppress the other. Every time a new fraud method is developed, a specially designed anti-fraud technique will come out soon. After that new fraud methods will keep coming out to avoid being detected. The second method in this paper has the potential of ending this circle. The method does not target any fraud attempts, but it observes the differences between user groups. As long as the fraudsters do not own related data of real user groups, it is almost impossible for fraudsters to avoid being detected by this method.

FoR codes (2008)




Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.