Information Extraction from Spam Emails using Stylistic and Semantic Features to Identify Spammers

Traditional anti spamming methods filter spam emails and prevent them from entering the inbox but take no measure to trace spammers and penalize them. We use natural language processing techniques to cluster spam emails from the same spammer based on the content and the style of the email. Spam emails from different sources are studied with features like stylistic, semantic and combination of both. Three sets of clustering are performed: clustering based on stylistic feature, clustering based on semantic feature and clustering based on combined feature. These clusters are then compared and evaluated. We notice that spam emails from the same sources have similarities and cluster together. These emails have URLs of the WebPages that the spammer is trying to promote. Clusters are mapped to the internet protocol (IP) of these URLs and the whois information of the IP addresses' help to get information about the source of spam.