Detection Method for User Click Fraud Based on Garbled Bloom Filter

EasyChair Preprint no. 735

Abstract

In order to solve the click fraud problem in the network, a detection method of user click fraud is presented with confusion Bloom filter. In this method, the user click fraud advertising set is found by the training data set attributes at first, and then use the Bloom filter and outlier mining algorithm for the detection and localization of suspected fraud. Finally, all suspected fraud members were classified by the Bias classification method to detect fraud. The experiment results show that this method can effectively shield visitors from accidental unconscious clicks, and significantly reduce the probability of click fraud.

Keyphrases: bias, Bloom filter, click fraud, Outlier Mining

@Booklet{EasyChair:735,
  author = {Chunliang Zhou and Le Wang},
  title = {Detection Method for User Click Fraud Based on Garbled Bloom Filter},
  howpublished = {EasyChair Preprint no. 735},

  year = {EasyChair, 2019}}