A mechanism to detect urdu Spam emails

Abstract

Electronic mail (Email) is being used for communication in all over the world as a basic communicational media. Few decades ago English was the only language of email but now almost facility of each language for email is present. Now people can use Urdu language as medium of Emails, social media and blog discussions. This Paper finds approaches of Spam/Ham categorization of English Emails are not able to categorize the Urdu emails because Urdu and English script are totally different in many aspects, so these are not directly applicable to Urdu. To categorization of emails into spam and ham is called Emails Spam filtering. Moreover, we need to develop tool by using machine learning technique that is appropriate in Urdu Language. In our work we have performed spam classification for Emails in Urdu language. We have collected different Spam and Ham Urdu Emails from different users. Most of the algorithms and techniques that are used for Spam classification in English and others languages are discussed and evaluated from different countries in this paper.