Electronic mail is one of today’s most important ways to communicate and transfer information. Because of fast delivery and easy to access, it is used almost in every aspect of communication in work and life. The continuous growth of email users has resulted in the increasing of unsolicited emails also known as Spam. SPAM email is well known problem for both corporate and personal users of email. Although SPAM has been well studied, both formally and informally, SPAM continues to be a significant problem. In current scenario, server side and client side anti spam filters are introduced for detecting different features of spam emails. So, Separation of spam from normal mails is essential. This paper surveys different spam filtering techniques. Techniques to separate spam mails are word based, content based, machine learning based and hybrid. Among them Machine learning techniques are most popular because of high accuracy and mathematical support. In this paper the overview of existing e-mail spam filtering methods is given. The classification, evaluation, and comparison of various machine learning-based methods are provided.