Combating against Web Spam through Content Features

Web spamming refers to use of unethical search engine optimization practices to gain better position on Search Engine Result Page (SERP). Making judgment on web-page to declare it as spam or ham is complicated issue because different search engines have different standards. Link-based spamming, cloaking and content spamming is main focus of different anti spam techniques. Even though these anti-spam techniques have had much success, however, these techniques still face problems when combating against a new kind of spamming techniques. This paper presents a usage of different machine learning methods which provides a solution for supervised classification problem. We have used WEBSPAM-UK-2007 public data set and in our experiments. The final results are compared and analyzed with well known classifiers. The results show that Jrip and J48 perform well compared to other two methods.