Unsolicited e-mail (Spam) has become a major issue for each e-mail user. In recent days it is very difficult to filter spam emails as these emails are written or generated in a very special way so that anti-spam filters cannot detect such emails. This Paper compares and discusses performance measures of certain categories of supervised machine learning techniques such as Bayes algorithms, lazy algorithms, tree algorithms, neural network, and support vector machines for classifying a spam e-mail corpus maintained by UCI Machine Learning Repository. The objective of this study is to consider the content of the emails, learn a finite dataset available and to develop a classification model that will able to predict whether an e-mail is spam or not.