An Automated Framework with Application to Study Url Based Online Advertisements Detection

A rapid growth of online advertisements results in unsolicited bulk of data being downloaded during web surfing. To tackle this problem a fast mechanism detecting adverts is required. In this paper we present the usefulness of URL based web-pages classification in the process of online advertisements detection. Our experiments are performed on seven popular classifiers using the real-life dataset obtained by human agents browsing the internet. We introduce a general and fully automated framework that allows us to do a comprehensive analysis by performing simultaneously hundreds of experiments. This study results in solution with 0.987 accuracy and 0.822 F-measure.