基于Web日志挖掘的欺骗点击检测模型

伴随着网络广告的高点击率而来的是大量的欺骗点击,如何准确识别出欺骗点击、保证用户的收益成为网络广告的当务之急。通过分析常用欺骗点击的手段、传统检测方法以及Web日志挖掘特点,提出了一个利用Web日志挖掘进行欺骗点击检测的模型。对该模型各个模块的实现方法进行了详细的阐述。实验表明,该模型简单高效,可以有效地检测出网络广告的欺骗点击。

A fraud-click detecting model based on Web log mining

There are a lot of fraud clicks following the high click rate of online advertising. How to accurately identify fraud clicks to ensure the user's online advertising revenue is becoming the top priority. By analyzing the means and features of fraud click and characteristics of Web log mining, we propose a fraud clicks detecting model based on Web log mining. The realization of each module is described in detail. Experiments show that the model is simple and efficient, it can effectively detect fraudulent clicks.