Abstract Rapid growth of the Internet has led to a proliferation of emails. Nowadays, it is common for an email user to receive tens or even hundreds of emails every day. To organize our emails so that they can be searched and maintained efficiently, we often group them into files. However, reading the emails one by one and filing them by hand is still a tedious process. Moreover, the problem is getting worse as the number of emails and folders keep increasing. Thus, the problem of automatic email classification is important and has gained much attention, especially in recent years. In this paper, we study the problem and focus on building a new proposed system used to classify emails automatically based on VCAD algorithm to help decision makers in educational institutions. Key Words: Email classification; Naive bayes; Support Vector Machines, k-nearest neighbor; Vector Cosine Angle Distance, Euclidean Distance .