A Survey on the Content Based Classification of E-mails using Classification Techniques

E-mail is one of the most widely used modes of written communication over the past two decades through the internet, also one of the greatest methods of communication that has been accepted for private message or vocation purpose of communication and its traffic has increased aggressively with the appearance of World Wide Web. So nowadays each and everyone have at least one e-mail account. Sometimes user receives the email consisting of same content repeatedly by multiple users. The main aim of this study is to analyze the performance of different pre - existing classification techniques and to select the best classification technique with more accuracy and efficiency, so that the e-mails consisting of same content known as duplicate e-mails can be classified and redirected to the original email, later forwarding them to the trash bin. Classification techniques are of two types – based on machine learning techniques and based on non – machine learning techniques. Based on machine learning techniques consist of the algorithms such as Naïve Bayesian Classifier, Support Vector Machine (SVM), K Nearest Neighbour (KNN) and many more. Based on non – machine learning techniques consist of Black/White List, Signatures, Keyword Checking, Mail Header Checking.