Abstract: Email communication is one of the fastest means of information sharing and has become successful among online users. This predominant success has made web users to generate anonymous contents which are called as spam. Research on identifying these fraudulent information has been a major research issue in recent years and continues to be a major threat. Spam occurs in the information in textual form, short messages and images. Variety of methods exists to ensure security like Naive Bayes, machine learning, Genetic algorithm and others. Machine learning techniques now days used to automatically filter the spam e-Mail in a very successful rate. Classifying the emails as genuine or vice versa is a major research concern. This study attempts to provide a study on this context and there by provide a framework for improving the security. Descriptions of the algorithms are presented and the comparison of their performance on the SpamAssassin spam corpus is presented.