Prevent XPath and CSS Based Scrapers by Using Markup Randomizer

摘要

Web Scraping may consider as data theft action, several researchers have introduced some approach es for addressing this issue. These solutions could solve the problem in partial ways and sometimes, solution cannot be applicable with modern web techniques. Consequently, in our work we have introduced a new approach for stopping web scraping in an efficient way and applicable with modern web techniques called Markup Randomizer, which changes the HTML and CSS in proper way randomly in timely manner. The best feature of our model is that each web page can use it without paying any efforts or restrictions in web site markup. Experiments done over collected dataset which consist of 30 websites divided into three categories: News, Currency Rates and Weather. The proposed model based on Markup Randomizer applied over this dataset. The aim of the experimental is to measure the Similarity, File Size and the time. During testing the proposed model, we get that a change on the markup done up to 50%, file size is changed and optimized after during the process. The required time to applying the model and generating the new markup is good and up to 2 minutes. Finally, we find that our proposed markup randomizer is accepted.