Author : Bhavini Kanoongo, Puja Jagani and Khushali Deulkar

Due to the accelerated advancement of the Internet, online advertisement plays an essential role in the advertising market. One of the present and conventional revenue models for online advertising comprises of charging for each click depending on the reputation of keywords and the number of contending advertisers. This pay-per-click model gives a chance to individuals or rival companies to generate false clicks (i.e., click fraud), which act as a serious problem to the evolution of healthy online advertising market. Click fraud is the intended clicking on advertisements with no actual interest in the product or the service offered. It is one of the most critical problems in online advertising. It is thus very important to build detection methods for click fraud. Using a combination of statistical measures, timing window concept and zero access auto clicking as well as search hijacking, we document the distinct methods used by every module, the architecture they use, and how they operate to dupe online advertisers.

Keywords: Data Mining, Fraud detection.

Article published in International Journal of Current Engineering and Technology, Vol.4, No.5 (Oct-2014)